You cannot select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
suricata/src/app-layer-parser.c

1201 lines
34 KiB
C

/* Copyright (C) 2007-2013 Open Information Security Foundation
*
* You can copy, redistribute or modify this Program under the terms of
* the GNU General Public License version 2 as published by the Free
* Software Foundation.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* version 2 along with this program; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
* 02110-1301, USA.
*/
/**
* \file
*
* \author Victor Julien <victor@inliniac.net>
*
* Generic App-layer parsing functions.
*/
#include "suricata-common.h"
#include "debug.h"
#include "util-unittest.h"
#include "decode.h"
#include "threads.h"
#include "util-print.h"
#include "util-pool.h"
#include "flow-util.h"
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
#include "flow-private.h"
#include "detect-engine-state.h"
#include "detect-engine-port.h"
#include "stream-tcp.h"
#include "stream-tcp-private.h"
#include "stream.h"
#include "stream-tcp-reassemble.h"
#include "app-layer.h"
#include "app-layer-protos.h"
#include "app-layer-parser.h"
#include "app-layer-smb.h"
#include "app-layer-smb2.h"
#include "app-layer-dcerpc.h"
#include "app-layer-dcerpc-udp.h"
#include "app-layer-htp.h"
#include "app-layer-ftp.h"
#include "app-layer-ssl.h"
#include "app-layer-ssh.h"
#include "app-layer-smtp.h"
#include "app-layer-dns-udp.h"
#include "app-layer-dns-tcp.h"
#include "conf.h"
#include "util-spm.h"
#include "util-debug.h"
#include "decode-events.h"
#include "util-unittest-helper.h"
#include "util-validate.h"
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
#include "runmodes.h"
typedef struct AppLayerParserThreadCtx_ {
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void *alproto_local_storage[FLOW_PROTO_MAX][ALPROTO_MAX];
} AppLayerParserThreadCtx;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
/**
* \brief App layer protocol parser context.
*/
typedef struct AppLayerParserpCtx_
{
/* 0 - to_server, 1 - to_client. */
int (*Parser[2])(Flow *f, void *protocol_state,
void *pstate,
uint8_t *input, uint32_t input_len,
void *local_storage);
char logger;
void *(*StateAlloc)(void);
void (*StateFree)(void *);
void (*StateTransactionFree)(void *, uint64_t);
void *(*LocalStorageAlloc)(void);
void (*LocalStorageFree)(void *);
void (*Truncate)(void *, uint8_t);
FileContainer *(*StateGetFiles)(void *, uint8_t);
AppLayerDecoderEvents *(*StateGetEvents)(void *, uint64_t);
int (*StateHasEvents)(void *);
int (*StateGetProgress)(void *alstate, uint8_t direction);
uint64_t (*StateGetTxCnt)(void *alstate);
void *(*StateGetTx)(void *alstate, uint64_t tx_id);
int (*StateGetProgressCompletionStatus)(uint8_t direction);
int (*StateGetEventInfo)(const char *event_name,
int *event_id, AppLayerEventType *event_type);
/* Indicates the direction the parser is ready to see the data
* the first time for a flow. Values accepted -
* STREAM_TOSERVER, STREAM_TOCLIENT */
uint8_t first_data_dir;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
#ifdef UNITTESTS
void (*RegisterUnittests)(void);
#endif
} AppLayerParserpCtx;
typedef struct AppLayerParserCtx_ {
AppLayerParserpCtx ctxs[FLOW_PROTO_MAX][ALPROTO_MAX];
} AppLayerParserCtx;
typedef struct AppLayerParserParserState_ {
uint8_t flags;
/* Indicates the current transaction that is being inspected.
* We have a var per direction. */
uint64_t inspect_id[2];
/* Indicates the current transaction being logged. Unlike inspect_id,
* we don't need a var per direction since we don't log a transaction
* unless we have the entire transaction. */
uint64_t log_id;
/* State version, incremented for each update. Can wrap around. */
uint16_t version;
/* Used to store decoder events. */
AppLayerDecoderEvents *decoder_events;
} AppLayerParserParserState;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
/* Static global version of the parser context.
* Post 2.0 let's look at changing this to move it out to app-layer.c. */
static AppLayerParserCtx alp_ctx;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
static void AppLayerParserTransactionsCleanup(uint16_t ipproto, AppProto alproto,
void *alstate, void *pstate)
{
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
AppLayerParserParserState *parser_state_store = pstate;
uint64_t inspect = 0, log = 0;
uint64_t min;
AppLayerParserpCtx *ctx = &alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto];
if (ctx->StateTransactionFree == NULL)
goto end;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
if (parser_state_store->inspect_id[0] < parser_state_store->inspect_id[1])
inspect = parser_state_store->inspect_id[0];
else
inspect = parser_state_store->inspect_id[1];
log = parser_state_store->log_id;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
if (ctx->logger == TRUE) {
min = log < inspect ? log : inspect;
if (min > 0)
ctx->StateTransactionFree(alstate, min - 1);
} else {
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
if (inspect > 0)
ctx->StateTransactionFree(alstate, inspect - 1);
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
end:
SCReturn;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void *AppLayerParserAllocAppLayerParserParserState(void)
{
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
AppLayerParserParserState *pstate = (AppLayerParserParserState *)SCMalloc(sizeof(*pstate));
if (pstate == NULL)
goto end;
memset(pstate, 0, sizeof(*pstate));
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
end:
SCReturnPtr(pstate, "pstate");
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void AppLayerParserDeAllocAppLayerParserParserState(void *pstate)
{
SCEnter();
if (((AppLayerParserParserState *)pstate)->decoder_events != NULL)
AppLayerDecoderEventsFreeEvents(((AppLayerParserParserState *)pstate)->decoder_events);
SCFree(pstate);
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturn;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
int AppLayerParserSetup(void)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
memset(&alp_ctx, 0, sizeof(alp_ctx));
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturnInt(0);
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
int AppLayerParserDeSetup(void)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturnInt(0);
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void *AppLayerParserGetCtxThread(void)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
AppProto i = 0;
int j = 0;
AppLayerParserThreadCtx *tctx;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
tctx = SCMalloc(sizeof(*tctx));
if (tctx == NULL)
goto end;
memset(tctx, 0, sizeof(*tctx));
for (i = 0; i < FLOW_PROTO_DEFAULT; i++) {
for (j = 0; j < ALPROTO_MAX; j++) {
tctx->alproto_local_storage[i][j] =
AppLayerParserGetProtocolParserLocalStorage(FlowGetReverseProtoMapping(i), j);
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
}
}
end:
SCReturnPtr(tctx, "void *");
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void AppLayerParserDestroyCtxThread(void *alpd_tctx)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
AppProto i = 0;
int j = 0;
AppLayerParserThreadCtx *tctx = (AppLayerParserThreadCtx *)alpd_tctx;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
for (i = 0; i < FLOW_PROTO_DEFAULT; i++) {
for (j = 0; j < ALPROTO_MAX; j++) {
AppLayerParserDestroyProtocolParserLocalStorage(FlowGetReverseProtoMapping(i),
j,
tctx->alproto_local_storage[i][j]);
}
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturn;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
int AppLayerParserConfParserEnabled(const char *ipproto,
const char *alproto_name)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCEnter();
int enabled = 1;
char param[100];
ConfNode *node;
int r;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
if (RunmodeIsUnittests())
goto enabled;
r = snprintf(param, sizeof(param), "%s%s%s", "app-layer.protocols.",
alproto_name, ".enabled");
if (r < 0) {
SCLogError(SC_ERR_FATAL, "snprintf failure.");
exit(EXIT_FAILURE);
} else if (r > (int)sizeof(param)) {
SCLogError(SC_ERR_FATAL, "buffer not big enough to write param.");
exit(EXIT_FAILURE);
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
node = ConfGetNode(param);
if (node == NULL) {
SCLogDebug("Entry for %s not found.", param);
r = snprintf(param, sizeof(param), "%s%s%s%s%s", "app-layer.protocols.",
alproto_name, ".", ipproto, ".enabled");
if (r < 0) {
SCLogError(SC_ERR_FATAL, "snprintf failure.");
exit(EXIT_FAILURE);
} else if (r > (int)sizeof(param)) {
SCLogError(SC_ERR_FATAL, "buffer not big enough to write param.");
exit(EXIT_FAILURE);
}
node = ConfGetNode(param);
if (node == NULL) {
SCLogDebug("Entry for %s not found.", param);
goto enabled;
}
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
if (strcasecmp(node->val, "yes") == 0) {
goto enabled;
} else if (strcasecmp(node->val, "no") == 0) {
goto disabled;
} else if (strcasecmp(node->val, "detection-only") == 0) {
goto enabled;
} else {
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCLogError(SC_ERR_FATAL, "Invalid value found for %s.", param);
exit(EXIT_FAILURE);
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
disabled:
enabled = 0;
enabled:
SCReturnInt(enabled);
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
/***** Parser related registration *****/
int AppLayerParserRegisterParser(uint16_t ipproto, AppProto alproto,
uint8_t direction,
int (*Parser)(Flow *f, void *protocol_state,
void *pstate,
uint8_t *buf, uint32_t buf_len,
void *local_storage))
{
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].
Parser[(direction & STREAM_TOSERVER) ? 0 : 1] = Parser;
SCReturnInt(0);
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void AppLayerParserRegisterParserAcceptableDataDirection(uint16_t ipproto, AppProto alproto,
uint8_t direction)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCEnter();
alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].first_data_dir |=
(direction & (STREAM_TOSERVER | STREAM_TOCLIENT));
SCReturn;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void AppLayerParserRegisterStateFuncs(uint16_t ipproto, AppProto alproto,
void *(*StateAlloc)(void),
void (*StateFree)(void *))
{
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].StateAlloc =
StateAlloc;
alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].StateFree =
StateFree;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturn;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void AppLayerParserRegisterLocalStorageFunc(uint16_t ipproto, AppProto alproto,
void *(*LocalStorageAlloc)(void),
void (*LocalStorageFree)(void *))
{
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].LocalStorageAlloc =
LocalStorageAlloc;
alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].LocalStorageFree =
LocalStorageFree;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturn;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void AppLayerParserRegisterGetFilesFunc(uint16_t ipproto, AppProto alproto,
FileContainer *(*StateGetFiles)(void *, uint8_t))
{
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].StateGetFiles =
StateGetFiles;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturn;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void AppLayerParserRegisterGetEventsFunc(uint16_t ipproto, AppProto alproto,
AppLayerDecoderEvents *(*StateGetEvents)(void *, uint64_t))
{
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].StateGetEvents =
StateGetEvents;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturn;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void AppLayerParserRegisterHasEventsFunc(uint16_t ipproto, AppProto alproto,
int (*StateHasEvents)(void *))
{
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].StateHasEvents =
StateHasEvents;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturn;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void AppLayerParserRegisterLogger(uint16_t ipproto, AppProto alproto)
{
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].logger = TRUE;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturn;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void AppLayerParserRegisterTruncateFunc(uint16_t ipproto, AppProto alproto,
void (*Truncate)(void *, uint8_t))
{
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].Truncate = Truncate;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturn;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void AppLayerParserRegisterGetStateProgressFunc(uint16_t ipproto, AppProto alproto,
int (*StateGetProgress)(void *alstate, uint8_t direction))
{
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].
StateGetProgress = StateGetProgress;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturn;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void AppLayerParserRegisterTxFreeFunc(uint16_t ipproto, AppProto alproto,
void (*StateTransactionFree)(void *, uint64_t))
{
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].
StateTransactionFree = StateTransactionFree;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturn;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void AppLayerParserRegisterGetTxCnt(uint16_t ipproto, AppProto alproto,
uint64_t (*StateGetTxCnt)(void *alstate))
{
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].
StateGetTxCnt = StateGetTxCnt;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturn;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void AppLayerParserRegisterGetTx(uint16_t ipproto, AppProto alproto,
void *(StateGetTx)(void *alstate, uint64_t tx_id))
{
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].
StateGetTx = StateGetTx;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturn;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void AppLayerParserRegisterGetStateProgressCompletionStatus(uint16_t ipproto,
uint16_t alproto,
int (*StateGetProgressCompletionStatus)(uint8_t direction))
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].
StateGetProgressCompletionStatus = StateGetProgressCompletionStatus;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturn;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void AppLayerParserRegisterGetEventInfo(uint16_t ipproto, AppProto alproto,
int (*StateGetEventInfo)(const char *event_name, int *event_id,
AppLayerEventType *event_type))
{
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].
StateGetEventInfo = StateGetEventInfo;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturn;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
/***** Get and transaction functions *****/
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void *AppLayerParserGetProtocolParserLocalStorage(uint16_t ipproto, AppProto alproto)
{
SCEnter();
if (alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].
LocalStorageAlloc != NULL)
{
SCReturnPtr(alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].
LocalStorageAlloc(), "void *");
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturnPtr(NULL, "void *");
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void AppLayerParserDestroyProtocolParserLocalStorage(uint16_t ipproto, AppProto alproto,
void *local_data)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
if (alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].
LocalStorageFree != NULL)
{
alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].
LocalStorageFree(local_data);
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturn;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
uint64_t AppLayerParserGetTransactionLogId(void *pstate)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturnCT(((AppLayerParserParserState *)pstate)->log_id, "uint64_t");
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void AppLayerParserSetTransactionLogId(void *pstate)
{
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
((AppLayerParserParserState *)pstate)->log_id++;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturn;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
uint64_t AppLayerParserGetTransactionInspectId(void *pstate, uint8_t direction)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturnCT(((AppLayerParserParserState *)pstate)->
inspect_id[direction & STREAM_TOSERVER ? 0 : 1], "uint64_t");
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void AppLayerParserSetTransactionInspectId(void *pstate,
uint16_t ipproto, AppProto alproto, void *alstate,
uint8_t direction)
{
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
uint8_t dir = (direction & STREAM_TOSERVER) ? 0 : 1;
uint64_t total_txs = AppLayerParserGetTxCnt(ipproto, alproto, alstate);
uint64_t idx = AppLayerParserGetTransactionInspectId(pstate, direction);
int state_done_progress = AppLayerParserGetStateProgressCompletionStatus(ipproto, alproto, direction);
void *tx;
int state_progress;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
for (; idx < total_txs; idx++) {
tx = AppLayerParserGetTx(ipproto, alproto, alstate, idx);
if (tx == NULL)
continue;
state_progress = AppLayerParserGetStateProgress(ipproto, alproto, tx, direction);
if (state_progress >= state_done_progress)
continue;
else
break;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
((AppLayerParserParserState *)pstate)->inspect_id[dir] = idx;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturn;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
AppLayerDecoderEvents *AppLayerParserGetDecoderEvents(void *pstate)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturnPtr(((AppLayerParserParserState *)pstate)->decoder_events,
"AppLayerDecoderEvents *");
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void AppLayerParserSetDecoderEvents(void *pstate, AppLayerDecoderEvents *devents)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
(((AppLayerParserParserState *)pstate)->decoder_events) = devents;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
AppLayerDecoderEvents *AppLayerParserGetEventsByTx(uint16_t ipproto, AppProto alproto,
void *alstate, uint64_t tx_id)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
AppLayerDecoderEvents *ptr = NULL;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
if (alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].
StateGetEvents != NULL)
{
ptr = alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].
StateGetEvents(alstate, tx_id);
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturnPtr(ptr, "AppLayerDecoderEvents *");
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
uint16_t AppLayerParserGetStateVersion(void *pstate)
{
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturnCT((pstate == NULL) ? 0 : ((AppLayerParserParserState *)pstate)->version,
"uint16_t");
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
FileContainer *AppLayerParserGetFiles(uint16_t ipproto, AppProto alproto,
void *alstate, uint8_t direction)
{
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
FileContainer *ptr = NULL;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
if (alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].
StateGetFiles != NULL)
{
ptr = alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].
StateGetFiles(alstate, direction);
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturnPtr(ptr, "FileContainer *");
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
int AppLayerParserGetStateProgress(uint16_t ipproto, AppProto alproto,
void *alstate, uint8_t direction)
{
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturnInt(alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].
StateGetProgress(alstate, direction));
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
uint64_t AppLayerParserGetTxCnt(uint16_t ipproto, AppProto alproto, void *alstate)
{
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturnCT(alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].
StateGetTxCnt(alstate), "uint64_t");
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void *AppLayerParserGetTx(uint16_t ipproto, AppProto alproto, void *alstate, uint64_t tx_id)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCEnter();
SCReturnPtr(alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].
StateGetTx(alstate, tx_id), "void *");
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
int AppLayerParserGetStateProgressCompletionStatus(uint16_t ipproto, AppProto alproto,
uint8_t direction)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCEnter();
SCReturnInt(alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].
StateGetProgressCompletionStatus(direction));
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
int AppLayerParserGetEventInfo(uint16_t ipproto, AppProto alproto, const char *event_name,
int *event_id, AppLayerEventType *event_type)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCEnter();
int ipproto_map = FlowGetProtoMapping(ipproto);
12 years ago
int r = (alp_ctx.ctxs[ipproto_map][alproto].StateGetEventInfo == NULL) ?
-1 : alp_ctx.ctxs[ipproto_map][alproto].StateGetEventInfo(event_name, event_id, event_type);
SCReturnInt(r);
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
uint8_t AppLayerParserGetFirstDataDir(uint16_t ipproto, uint16_t alproto)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCEnter();
SCReturnCT(alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].
first_data_dir, "uint8_t");
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
uint64_t AppLayerParserGetTransactionActive(uint16_t ipproto, AppProto alproto, void *pstate, uint8_t direction)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCEnter();
AppLayerParserParserState *state = (AppLayerParserParserState *)pstate;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
uint64_t active_id;
uint64_t log_id = state->log_id;
uint64_t inspect_id = state->inspect_id[direction & STREAM_TOSERVER ? 0 : 1];
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
if (alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].logger == TRUE) {
active_id = (log_id < inspect_id) ? log_id : inspect_id;
} else {
active_id = inspect_id;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturnCT(active_id, "uint64_t");
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
/***** General *****/
int AppLayerParserParse(void *tctx, Flow *f, AppProto alproto,
uint8_t flags, uint8_t *input, uint32_t input_len)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
AppLayerParserParserState *pstate = NULL;
AppLayerParserpCtx *p = &alp_ctx.ctxs[FlowGetProtoMapping(f->proto)][alproto];
TcpSession *ssn = NULL;
void *alstate = NULL;
AppLayerParserThreadCtx *alp_tctx = (AppLayerParserThreadCtx *)tctx;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
/* we don't have the parser registered for this protocol */
if (p->StateAlloc == NULL)
goto end;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
/* Used only if it's TCP */
ssn = f->protoctx;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
/* Do this check before calling AppLayerParse */
if (flags & STREAM_GAP) {
SCLogDebug("stream gap detected (missing packets), "
"this is not yet supported.");
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
if (f->alstate != NULL)
AppLayerParserStreamTruncated(f->proto, alproto, f->alstate, flags);
goto error;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
/* Get the parser state (if any) */
pstate = f->alparser;
if (pstate == NULL) {
f->alparser = pstate = AppLayerParserAllocAppLayerParserParserState();
if (pstate == NULL)
goto error;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
pstate->version++;
SCLogDebug("app layer parser state version incremented to %"PRIu16,
pstate->version);
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
if (flags & STREAM_EOF)
AppLayerParserParserStateSetFlag(pstate, APP_LAYER_PARSER_EOF);
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
alstate = f->alstate;
if (alstate == NULL) {
f->alstate = alstate = p->StateAlloc();
if (alstate == NULL)
goto error;
SCLogDebug("alloced new app layer state %p (name %s)",
alstate, AppLayerGetProtoName(f->alproto));
} else {
SCLogDebug("using existing app layer state %p (name %s))",
alstate, AppLayerGetProtoName(f->alproto));
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
/* invoke the recursive parser, but only on data. We may get empty msgs on EOF */
if (input_len > 0) {
/* invoke the parser */
if (p->Parser[(flags & STREAM_TOSERVER) ? 0 : 1](f, alstate, pstate,
input, input_len,
alp_tctx->alproto_local_storage[FlowGetProtoMapping(f->proto)][alproto]) < 0)
{
goto error;
}
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
/* set the packets to no inspection and reassembly if required */
if (pstate->flags & APP_LAYER_PARSER_NO_INSPECTION) {
AppLayerParserSetEOF(pstate);
FlowSetNoPayloadInspectionFlag(f);
FlowSetSessionNoApplayerInspectionFlag(f);
/* Set the no reassembly flag for both the stream in this TcpSession */
if (pstate->flags & APP_LAYER_PARSER_NO_REASSEMBLY) {
if (ssn != NULL) {
StreamTcpSetSessionNoReassemblyFlag(ssn,
flags & STREAM_TOCLIENT ? 1 : 0);
StreamTcpSetSessionNoReassemblyFlag(ssn,
flags & STREAM_TOSERVER ? 1 : 0);
}
}
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
/* next, see if we can get rid of transactions now */
AppLayerParserTransactionsCleanup(f->proto, alproto, alstate, pstate);
/* stream truncated, inform app layer */
if (flags & STREAM_DEPTH)
AppLayerParserStreamTruncated(f->proto, alproto, alstate, flags);
end:
SCReturnInt(0);
error:
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
if (ssn != NULL) {
/* Set the no app layer inspection flag for both
* the stream in this Flow */
FlowSetSessionNoApplayerInspectionFlag(f);
AppLayerParserSetEOF(pstate);
}
SCReturnInt(-1);
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void AppLayerParserSetEOF(void *pstate)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
if (pstate == NULL)
goto end;
AppLayerParserParserStateSetFlag(pstate, APP_LAYER_PARSER_EOF);
/* increase version so we will inspect it one more time
* with the EOF flags now set */
((AppLayerParserParserState *)pstate)->version++;
end:
SCReturn;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
int AppLayerParserHasDecoderEvents(uint16_t ipproto, AppProto alproto,
void *alstate, void *pstate,
uint8_t flags)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
if (alstate == NULL || pstate == NULL)
goto not_present;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
AppLayerDecoderEvents *decoder_events;
uint64_t tx_id;
uint64_t max_id;
if (AppLayerParserProtocolIsTxEventAware(ipproto, alproto)) {
/* fast path if supported by alproto */
if (alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].StateHasEvents != NULL) {
if (alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].
StateHasEvents(alstate) == 1)
{
goto present;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
} else {
/* check each tx */
tx_id = AppLayerParserGetTransactionInspectId(pstate, flags);
max_id = AppLayerParserGetTxCnt(ipproto, alproto, alstate);
for ( ; tx_id < max_id; tx_id++) {
decoder_events = AppLayerParserGetEventsByTx(ipproto, alproto, alstate, tx_id);
if (decoder_events && decoder_events->cnt)
goto present;
}
}
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
decoder_events = AppLayerParserGetDecoderEvents(pstate);
if (decoder_events && decoder_events->cnt)
goto present;
/* if we have reached here, we don't have events */
not_present:
SCReturnInt(0);
present:
SCReturnInt(1);
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
int AppLayerParserProtocolIsTxEventAware(uint16_t ipproto, AppProto alproto)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCEnter();
int ipproto_map = FlowGetProtoMapping(ipproto);
12 years ago
int r = (alp_ctx.ctxs[ipproto_map][alproto].StateHasEvents == NULL) ? 0 : 1;
SCReturnInt(r);
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
int AppLayerParserProtocolSupportsTxs(uint16_t ipproto, AppProto alproto)
{
SCEnter();
int ipproto_map = FlowGetProtoMapping(ipproto);
12 years ago
int r = (alp_ctx.ctxs[ipproto_map][alproto].StateTransactionFree == NULL) ? 0 : 1;
SCReturnInt(r);
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void AppLayerParserTriggerRawStreamReassembly(Flow *f)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
if (f != NULL && f->protoctx != NULL)
StreamTcpReassembleTriggerRawReassembly(f->protoctx);
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturn;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
/***** Cleanup *****/
void AppLayerParserCleanupParserState(uint16_t ipproto, AppProto alproto, void *alstate, void *pstate)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
AppLayerParserpCtx *ctx = &alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto];
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
if (ctx->StateFree != NULL && alstate != NULL)
ctx->StateFree(alstate);
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
/* free the app layer parser api state */
if (pstate != NULL)
AppLayerParserDeAllocAppLayerParserParserState(pstate);
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturn;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void AppLayerParserRegisterProtocolParsers(void)
{
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
RegisterHTPParsers();
RegisterSSLParsers();
RegisterSMBParsers();
/** \todo bug 719 */
//RegisterSMB2Parsers();
RegisterDCERPCParsers();
RegisterDCERPCUDPParsers();
RegisterFTPParsers();
/* we are disabling the ssh parser temporarily, since we are moving away
* from some of the archaic features we use in the app layer. We will
* reintroduce this parser. Also do note that keywords that rely on
* the ssh parser would now be disabled */
#if 0
RegisterSSHParsers();
#endif
RegisterSMTPParsers();
RegisterDNSUDPParsers();
RegisterDNSTCPParsers();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
/** IMAP */
AppLayerProtoDetectRegisterProtocol(ALPROTO_IMAP, "imap");
if (AppLayerProtoDetectConfProtoDetectionEnabled("tcp", "imap")) {
if (AppLayerProtoDetectPMRegisterPatternCS(IPPROTO_TCP, ALPROTO_IMAP,
"1|20|capability", 12, 0, STREAM_TOSERVER) < 0)
{
SCLogInfo("imap proto registration failure\n");
exit(EXIT_FAILURE);
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
} else {
SCLogInfo("Protocol detection and parser disabled for %s protocol.",
"imap");
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
/** MSN Messenger */
AppLayerProtoDetectRegisterProtocol(ALPROTO_MSN, "msn");
if (AppLayerProtoDetectConfProtoDetectionEnabled("tcp", "msn")) {
if (AppLayerProtoDetectPMRegisterPatternCS(IPPROTO_TCP, ALPROTO_MSN,
"msn", 10, 6, STREAM_TOSERVER) < 0)
{
SCLogInfo("msn proto registration failure\n");
exit(EXIT_FAILURE);
}
} else {
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCLogInfo("Protocol detection and parser disabled for %s protocol.",
"msn");
}
return;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void AppLayerParserParserStateSetFlag(void *pstate, uint8_t flag)
{
SCEnter();
((AppLayerParserParserState *)pstate)->flags |= flag;
SCReturn;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
int AppLayerParserParserStateIssetFlag(void *pstate, uint8_t flag)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCEnter();
SCReturnInt(((AppLayerParserParserState *)pstate)->flags & flag);
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void AppLayerParserStreamTruncated(uint16_t ipproto, AppProto alproto, void *alstate,
uint8_t direction)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
if (alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].Truncate != NULL)
alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].Truncate(alstate, direction);
SCReturn;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
#ifdef DEBUG
void AppLayerParserPrintDetailsParserState(void *pstate)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
if (pstate == NULL)
SCReturn;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
AppLayerParserParserState *p = (AppLayerParserParserState *)pstate;
SCLogDebug("AppLayerParser parser state information for parser state p(%p). "
"p->inspect_id[0](%"PRIu64"), "
"p->inspect_id[1](%"PRIu64"), "
"p->log_id(%"PRIu64"), "
"p->version(%"PRIu16"), "
"p->decoder_events(%p).",
pstate, p->inspect_id[0], p->inspect_id[1], p->log_id,
p->version, p->decoder_events);
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturn;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
#endif
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
/***** Unittests *****/
#ifdef UNITTESTS
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
static AppLayerParserCtx alp_ctx_backup_unittest;
typedef struct TestState_ {
uint8_t test;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
} TestState;
/**
* \brief Test parser function to test the memory deallocation of app layer
* parser of occurence of an error.
*/
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
static int TestProtocolParser(Flow *f, void *test_state, void *pstate,
uint8_t *input, uint32_t input_len,
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void *local_data)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCEnter();
SCReturnInt(-1);
}
/** \brief Function to allocates the Test protocol state memory
*/
static void *TestProtocolStateAlloc(void)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCEnter();
void *s = SCMalloc(sizeof(TestState));
if (unlikely(s == NULL))
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
goto end;
memset(s, 0, sizeof(TestState));
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
end:
SCReturnPtr(s, "TestState");
}
/** \brief Function to free the Test Protocol state memory
*/
static void TestProtocolStateFree(void *s)
{
SCFree(s);
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void AppLayerParserRegisterProtocolUnittests(uint16_t ipproto, AppProto alproto,
void (*RegisterUnittests)(void))
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCEnter();
alp_ctx.ctxs[FlowGetProtoMapping(ipproto)][alproto].
RegisterUnittests = RegisterUnittests;
SCReturn;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void AppLayerParserBackupParserTable(void)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCEnter();
alp_ctx_backup_unittest = alp_ctx;
memset(&alp_ctx, 0, sizeof(alp_ctx));
SCReturn;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void AppLayerParserRestoreParserTable(void)
{
SCEnter();
alp_ctx = alp_ctx_backup_unittest;
memset(&alp_ctx_backup_unittest, 0, sizeof(alp_ctx_backup_unittest));
SCReturn;
}
/**
* \test Test the deallocation of app layer parser memory on occurance of
* error in the parsing process.
*/
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
static int AppLayerParserTest01(void)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
AppLayerParserBackupParserTable();
int result = 0;
Flow *f = NULL;
uint8_t testbuf[] = { 0x11 };
uint32_t testlen = sizeof(testbuf);
TcpSession ssn;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void *alp_tctx = AppLayerParserGetCtxThread();
memset(&ssn, 0, sizeof(ssn));
/* Register the Test protocol state and parser functions */
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
AppLayerParserRegisterParser(IPPROTO_TCP, ALPROTO_TEST, STREAM_TOSERVER,
TestProtocolParser);
AppLayerParserRegisterStateFuncs(IPPROTO_TCP, ALPROTO_TEST,
TestProtocolStateAlloc, TestProtocolStateFree);
f = UTHBuildFlow(AF_INET, "1.2.3.4", "4.3.2.1", 20, 40);
if (f == NULL)
goto end;
f->protoctx = &ssn;
f->alproto = ALPROTO_TEST;
f->proto = IPPROTO_TCP;
StreamTcpInitConfig(TRUE);
SCMutexLock(&f->m);
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
int r = AppLayerParserParse(alp_tctx, f, ALPROTO_TEST, STREAM_TOSERVER|STREAM_EOF,
testbuf, testlen);
if (r != -1) {
printf("returned %" PRId32 ", expected -1: ", r);
SCMutexUnlock(&f->m);
goto end;
}
SCMutexUnlock(&f->m);
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
if (!(f->flags & FLOW_NO_APPLAYER_INSPECTION)) {
printf("flag should have been set, but is not: ");
goto end;
}
result = 1;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
end:
AppLayerParserRestoreParserTable();
StreamTcpFreeConfig(TRUE);
UTHFreeFlow(f);
return result;
}
/**
* \test Test the deallocation of app layer parser memory on occurance of
* error in the parsing process for UDP.
*/
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
static int AppLayerParserTest02(void)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
AppLayerParserBackupParserTable();
int result = 1;
Flow *f = NULL;
uint8_t testbuf[] = { 0x11 };
uint32_t testlen = sizeof(testbuf);
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void *alp_tctx = AppLayerParserGetCtxThread();
/* Register the Test protocol state and parser functions */
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
AppLayerParserRegisterParser(IPPROTO_UDP, ALPROTO_TEST, STREAM_TOSERVER,
TestProtocolParser);
AppLayerParserRegisterStateFuncs(IPPROTO_UDP, ALPROTO_TEST,
TestProtocolStateAlloc, TestProtocolStateFree);
f = UTHBuildFlow(AF_INET, "1.2.3.4", "4.3.2.1", 20, 40);
if (f == NULL)
goto end;
f->alproto = ALPROTO_TEST;
f->proto = IPPROTO_UDP;
StreamTcpInitConfig(TRUE);
SCMutexLock(&f->m);
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
int r = AppLayerParserParse(alp_tctx, f, ALPROTO_TEST, STREAM_TOSERVER|STREAM_EOF, testbuf,
testlen);
if (r != -1) {
printf("returned %" PRId32 ", expected -1: \n", r);
result = 0;
SCMutexUnlock(&f->m);
goto end;
}
SCMutexUnlock(&f->m);
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
end:
AppLayerParserRestoreParserTable();
StreamTcpFreeConfig(TRUE);
UTHFreeFlow(f);
return result;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
void AppLayerParserRegisterUnittests(void)
{
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCEnter();
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
int ip;
uint16_t alproto;
AppLayerParserpCtx *ctx;
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
for (ip = 0; ip < FLOW_PROTO_DEFAULT; ip++) {
for (alproto = 0; alproto < ALPROTO_MAX; alproto++) {
ctx = &alp_ctx.ctxs[ip][alproto];
if (ctx->RegisterUnittests == NULL)
continue;
ctx->RegisterUnittests();
}
}
UtRegisterTest("AppLayerParserTest01", AppLayerParserTest01, 1);
UtRegisterTest("AppLayerParserTest02", AppLayerParserTest02, 1);
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
SCReturn;
}
App layer API rewritten. The main files in question are: app-layer.[ch], app-layer-detect-proto.[ch] and app-layer-parser.[ch]. Things addressed in this commit: - Brings out a proper separation between protocol detection phase and the parser phase. - The dns app layer now is registered such that we don't use "dnstcp" and "dnsudp" in the rules. A user who previously wrote a rule like this - "alert dnstcp....." or "alert dnsudp....." would now have to use, alert dns (ipproto:tcp;) or alert udp (app-layer-protocol:dns;) or alert ip (ipproto:udp; app-layer-protocol:dns;) The same rules extend to other another such protocol, dcerpc. - The app layer parser api now takes in the ipproto while registering callbacks. - The app inspection/detection engine also takes an ipproto. - All app layer parser functions now take direction as STREAM_TOSERVER or STREAM_TOCLIENT, as opposed to 0 or 1, which was taken by some of the functions. - FlowInitialize() and FlowRecycle() now resets proto to 0. This is needed by unittests, which would try to clean the flow, and that would call the api, AppLayerParserCleanupParserState(), which would try to clean the app state, but the app layer now needs an ipproto to figure out which api to internally call to clean the state, and if the ipproto is 0, it would return without trying to clean the state. - A lot of unittests are now updated where if they are using a flow and they need to use the app layer, we would set a flow ipproto. - The "app-layer" section in the yaml conf has also been updated as well.
12 years ago
#endif