From 6fc0c97ab284021b72e3bc962f7fa879ffcad65b Mon Sep 17 00:00:00 2001
From: Josip Sokcevic <sokcevic@chromium.org>
Date: Mon, 22 Jan 2024 21:15:40 +0000
Subject: [PATCH] [git_cache] Track if git cache is fully initialized

git_cache populate can get interrupted midway (e.g. by LUCI CV sending a
signal that build is no longer needed). When that happens, a git mirror
may be in a state where some commits are available, but cloning such
repositry results in an empty repsitory.

This change ensures that the initial fetch operation finished
successfully.

R=ddoman@google.com

Bug: 1517944
Change-Id: I1ee860860877dbfff7a444b45fe4515fe26b248c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/tools/depot_tools/+/5219032
Commit-Queue: Josip Sokcevic <sokcevic@chromium.org>
Reviewed-by: Scott Lee <ddoman@chromium.org>
---
 git_cache.py            | 37 ++++++++++++++++++++++++++++---------
 tests/git_cache_test.py | 28 ++++++++++++++++++++++++++++
 2 files changed, 56 insertions(+), 9 deletions(-)

diff --git a/git_cache.py b/git_cache.py
index 204e59195..0c7c92487 100755
--- a/git_cache.py
+++ b/git_cache.py
@@ -26,6 +26,7 @@ import subcommand
 GC_AUTOPACKLIMIT = 50
 
 GIT_CACHE_CORRUPT_MESSAGE = 'WARNING: The Git cache is corrupt.'
+INIT_SENTIENT_FILE = ".mirror_init"
 
 # gsutil creates many processes and threads. Creating too many gsutil cp
 # processes may result in running out of resources, and may perform worse due to
@@ -136,6 +137,10 @@ class Mirror(object):
             self.print('%s took %.1f minutes' % (what,
                                                  (time.time() - start) / 60.0))
 
+    @property
+    def _init_sentient_file(self):
+        return os.path.join(self.mirror_path, INIT_SENTIENT_FILE)
+
     @property
     def bootstrap_bucket(self):
         b = os.getenv('OVERRIDE_BOOTSTRAP_BUCKET')
@@ -459,6 +464,9 @@ class Mirror(object):
                 # 2. Project doesn't have a bootstrap folder.
                 # Start with a bare git dir.
                 self.RunGit(['init', '--bare'])
+                with open(self._init_sentient_file, 'w'):
+                    # Create sentient file
+                    pass
                 # Set appropriate symbolic-ref
                 remote_info = exponential_backoff_retry(
                     lambda: subprocess.check_output(
@@ -523,6 +531,8 @@ class Mirror(object):
                     self.RunGit(['fetch', 'origin', commit], retry=True)
             except subprocess.CalledProcessError:
                 logging.warning('Fetch of %s failed' % commit)
+        if os.path.isfile(self._init_sentient_file):
+            os.remove(self._init_sentient_file)
 
     def populate(self,
                  depth=None,
@@ -537,20 +547,29 @@ class Mirror(object):
             depth = 10000
         gclient_utils.safe_makedirs(self.GetCachePath())
 
+        def bootstrap(force=False):
+            self._ensure_bootstrapped(depth,
+                                      bootstrap,
+                                      reset_fetch_config,
+                                      force=force)
+            self._fetch(verbose, depth, no_fetch_tags, reset_fetch_config)
+
+        def wipe_cache():
+            self.print(GIT_CACHE_CORRUPT_MESSAGE)
+            gclient_utils.rmtree(self.mirror_path)
+
         with lockfile.lock(self.mirror_path, lock_timeout):
+            if os.path.isfile(self._init_sentient_file):
+                # Previous bootstrap didn't finish
+                wipe_cache()
+
             try:
-                self._ensure_bootstrapped(depth, bootstrap, reset_fetch_config)
-                self._fetch(verbose, depth, no_fetch_tags, reset_fetch_config)
+                bootstrap()
             except ClobberNeeded:
                 # This is a major failure, we need to clean and force a
                 # bootstrap.
-                gclient_utils.rmtree(self.mirror_path)
-                self.print(GIT_CACHE_CORRUPT_MESSAGE)
-                self._ensure_bootstrapped(depth,
-                                          bootstrap,
-                                          reset_fetch_config,
-                                          force=True)
-                self._fetch(verbose, depth, no_fetch_tags, reset_fetch_config)
+                wipe_cache()
+                bootstrap(force=True)
 
     def update_bootstrap(self, prune=False, gc_aggressive=False):
         # NOTE: There have been cases where repos were being recursively
diff --git a/tests/git_cache_test.py b/tests/git_cache_test.py
index cd7851ec3..b9a8d0feb 100755
--- a/tests/git_cache_test.py
+++ b/tests/git_cache_test.py
@@ -148,6 +148,34 @@ class GitCacheTest(unittest.TestCase):
         self.assertNotIn(git_cache.GIT_CACHE_CORRUPT_MESSAGE,
                          sys.stdout.getvalue())
 
+    @mock.patch('sys.stdout', StringIO())
+    def testBadInit(self):
+        self.git(['init', '-q'])
+        with open(os.path.join(self.origin_dir, 'foo'), 'w') as f:
+            f.write('touched\n')
+        self.git(['add', 'foo'])
+        self.git([
+            '-c', 'user.name=Test user', '-c', 'user.email=joj@test.com',
+            'commit', '-m', 'foo'
+        ])
+
+        mirror = git_cache.Mirror(self.origin_dir)
+
+        # Simulate init being interrupted during fetch phase.
+        with mock.patch.object(mirror, '_fetch'):
+            mirror.populate()
+
+        # Corrupt message is not expected at this point since it was
+        # "interrupted".
+        self.assertNotIn(git_cache.GIT_CACHE_CORRUPT_MESSAGE,
+                         sys.stdout.getvalue())
+
+        # We call mirror.populate() without _fetch patched. This time, a
+        # sentient file should prompt cache deletion.
+        mirror.populate()
+        self.assertIn(git_cache.GIT_CACHE_CORRUPT_MESSAGE,
+                      sys.stdout.getvalue())
+
     def _makeGitRepoWithTag(self):
         self.git(['init', '-q'])
         with open(os.path.join(self.origin_dir, 'foo'), 'w') as f: