revlog: add a mechanism to verify expected file position before appending
If someone uses `hg debuglocks`, or some non-hg process writes to the .hg
directory without respecting the locks, or if the repo's on a networked
filesystem, it's possible for the revlog code to write out corrupted data.
The form of this corruption can vary depending on what data was written and how
that happened. We are in the "networked filesystem" case (though I've had users
also do this to themselves with the "`hg debuglocks`" scenario), and most often
see this with the changelog. What ends up happening is we produce two items
(let's call them rev1 and rev2) in the .i file that have the same linkrev,
baserev, and offset into the .d file, while the data in the .d file is appended
properly. rev2's compressed_size is accurate for rev2, but when we go to
decompress the data in the .d file, we use the offset that's recorded in the
index file, which is the same as rev1, and attempt to decompress
rev2.compressed_size bytes of rev1's data. This usually does not succeed. :)
When using inline data, this also fails, though I haven't investigated why too
closely. This shows up as a "patch decode" error. I believe what's happening
there is that we're basically ignoring the offset field, getting the data
properly, but since baserev != rev, it thinks this is a delta based on rev
(instead of a full text) and can't actually apply it as such.
For now, I'm going to make this an optional component and default it to entirely
off. I may increase the default severity of this in the future, once I've
enabled it for my users and we gain more experience with it. Luckily, most of my
users have a versioned filesystem and can roll back to before the corruption has
been written, it's just a hassle to do so and not everyone knows how (so it's a
support burden). Users on other filesystems will not have that luxury, and this
can cause them to have a corrupted repository that they are unlikely to know how
to resolve, and they'll see this as a data-loss event. Refusing to create the
corruption is a much better user experience.
This mechanism is not perfect. There may be false-negatives (racy writes that
are not detected). There should not be any false-positives (non-racy writes that
are detected as such). This is not a mechanism that makes putting a repo on a
networked filesystem "safe" or "supported", just *less* likely to cause
corruption.
Differential Revision: https://phab.mercurial-scm.org/D9952
--- a/hgext/git/__init__.py Tue Feb 23 22:58:30 2021 -0800
+++ b/hgext/git/__init__.py Wed Feb 03 16:33:10 2021 -0800
@@ -90,7 +90,7 @@
return os.path.join(self.path, b'..', b'.hg', f)
raise NotImplementedError(b'Need to pick file for %s.' % f)
- def changelog(self, trypending):
+ def changelog(self, trypending, concurrencychecker):
# TODO we don't have a plan for trypending in hg's git support yet
return gitlog.changelog(self.git, self._db)
--- a/mercurial/changelog.py Tue Feb 23 22:58:30 2021 -0800
+++ b/mercurial/changelog.py Wed Feb 03 16:33:10 2021 -0800
@@ -380,7 +380,7 @@
class changelog(revlog.revlog):
- def __init__(self, opener, trypending=False):
+ def __init__(self, opener, trypending=False, concurrencychecker=None):
"""Load a changelog revlog using an opener.
If ``trypending`` is true, we attempt to load the index from a
@@ -389,6 +389,9 @@
revision) data for a transaction that hasn't been finalized yet.
It exists in a separate file to facilitate readers (such as
hooks processes) accessing data before a transaction is finalized.
+
+ ``concurrencychecker`` will be passed to the revlog init function, see
+ the documentation there.
"""
if trypending and opener.exists(b'00changelog.i.a'):
indexfile = b'00changelog.i.a'
@@ -404,6 +407,7 @@
checkambig=True,
mmaplargeindex=True,
persistentnodemap=opener.options.get(b'persistent-nodemap', False),
+ concurrencychecker=concurrencychecker,
)
if self._initempty and (self.version & 0xFFFF == revlog.REVLOGV1):
--- a/mercurial/configitems.py Tue Feb 23 22:58:30 2021 -0800
+++ b/mercurial/configitems.py Wed Feb 03 16:33:10 2021 -0800
@@ -580,6 +580,11 @@
default=0,
)
coreconfigitem(
+ b'debug',
+ b'revlog.verifyposition.changelog',
+ default=b'',
+)
+coreconfigitem(
b'defaults',
b'.*',
default=None,
--- a/mercurial/localrepo.py Tue Feb 23 22:58:30 2021 -0800
+++ b/mercurial/localrepo.py Wed Feb 03 16:33:10 2021 -0800
@@ -84,7 +84,10 @@
stringutil,
)
-from .revlogutils import constants as revlogconst
+from .revlogutils import (
+ concurrency_checker as revlogchecker,
+ constants as revlogconst,
+)
release = lockmod.release
urlerr = util.urlerr
@@ -1639,7 +1642,10 @@
def changelog(self):
# load dirstate before changelog to avoid race see issue6303
self.dirstate.prefetch_parents()
- return self.store.changelog(txnutil.mayhavepending(self.root))
+ return self.store.changelog(
+ txnutil.mayhavepending(self.root),
+ concurrencychecker=revlogchecker.get_checker(self.ui, b'changelog'),
+ )
@storecache(b'00manifest.i')
def manifestlog(self):
--- a/mercurial/revlog.py Tue Feb 23 22:58:30 2021 -0800
+++ b/mercurial/revlog.py Wed Feb 03 16:33:10 2021 -0800
@@ -421,6 +421,11 @@
If `upperboundcomp` is not None, this is the expected maximal gain from
compression for the data content.
+
+ `concurrencychecker` is an optional function that receives 3 arguments: a
+ file handle, a filename, and an expected position. It should check whether
+ the current position in the file handle is valid, and log/warn/fail (by
+ raising).
"""
_flagserrorclass = error.RevlogError
@@ -435,6 +440,7 @@
censorable=False,
upperboundcomp=None,
persistentnodemap=False,
+ concurrencychecker=None,
):
"""
create a revlog object
@@ -490,6 +496,8 @@
self._loadindex()
+ self._concurrencychecker = concurrencychecker
+
def _loadindex(self):
mmapindexthreshold = None
opts = self.opener.options
@@ -2284,6 +2292,21 @@
curr = len(self)
prev = curr - 1
offset = self.end(prev)
+
+ if self._concurrencychecker:
+ if self._inline:
+ # offset is "as if" it were in the .d file, so we need to add on
+ # the size of the entry metadata.
+ self._concurrencychecker(
+ ifh, self.indexfile, offset + curr * self._io.size
+ )
+ else:
+ # Entries in the .i are a consistent size.
+ self._concurrencychecker(
+ ifh, self.indexfile, curr * self._io.size
+ )
+ self._concurrencychecker(dfh, self.datafile, offset)
+
p1r, p2r = self.rev(p1), self.rev(p2)
# full versions are inserted when the needed deltas
--- /dev/null Thu Jan 01 00:00:00 1970 +0000
+++ b/mercurial/revlogutils/concurrency_checker.py Wed Feb 03 16:33:10 2021 -0800
@@ -0,0 +1,38 @@
+from ..i18n import _
+from .. import error
+
+
+def get_checker(ui, revlog_name=b'changelog'):
+ """Get a function that checks file handle position is as expected.
+
+ This is used to ensure that files haven't been modified outside of our
+ knowledge (such as on a networked filesystem, if `hg debuglocks` was used,
+ or writes to .hg that ignored locks happened).
+
+ Due to revlogs supporting a concept of buffered, delayed, or diverted
+ writes, we're allowing the files to be shorter than expected (the data may
+ not have been written yet), but they can't be longer.
+
+ Please note that this check is not perfect; it can't detect all cases (there
+ may be false-negatives/false-OKs), but it should never claim there's an
+ issue when there isn't (false-positives/false-failures).
+ """
+
+ vpos = ui.config(b'debug', b'revlog.verifyposition.' + revlog_name)
+ # Avoid any `fh.tell` cost if this isn't enabled.
+ if not vpos or vpos not in [b'log', b'warn', b'fail']:
+ return None
+
+ def _checker(fh, fn, expected):
+ if fh.tell() <= expected:
+ return
+
+ msg = _(b'%s: file cursor at position %d, expected %d')
+ # Always log if we're going to warn or fail.
+ ui.log(b'debug', msg + b'\n', fn, fh.tell(), expected)
+ if vpos == b'warn':
+ ui.warn((msg + b'\n') % (fn, fh.tell(), expected))
+ elif vpos == b'fail':
+ raise error.RevlogError(msg % (fn, fh.tell(), expected))
+
+ return _checker
--- a/mercurial/store.py Tue Feb 23 22:58:30 2021 -0800
+++ b/mercurial/store.py Wed Feb 03 16:33:10 2021 -0800
@@ -433,8 +433,12 @@
l.sort()
return l
- def changelog(self, trypending):
- return changelog.changelog(self.vfs, trypending=trypending)
+ def changelog(self, trypending, concurrencychecker=None):
+ return changelog.changelog(
+ self.vfs,
+ trypending=trypending,
+ concurrencychecker=concurrencychecker,
+ )
def manifestlog(self, repo, storenarrowmatch):
rootstore = manifest.manifestrevlog(self.vfs)
--- /dev/null Thu Jan 01 00:00:00 1970 +0000
+++ b/tests/test-racy-mutations.t Wed Feb 03 16:33:10 2021 -0800
@@ -0,0 +1,102 @@
+#testcases skip-detection fail-if-detected
+
+Test situations that "should" only be reproducible:
+- on networked filesystems, or
+- user using `hg debuglocks` to eliminate the lock file, or
+- something (that doesn't respect the lock file) writing to the .hg directory
+while we're running
+
+ $ hg init a
+ $ cd a
+
+ $ cat > "$TESTTMP/waitlock_editor.sh" <<EOF
+ > [ -n "\${WAITLOCK_ANNOUNCE:-}" ] && touch "\${WAITLOCK_ANNOUNCE}"
+ > f="\${WAITLOCK_FILE}"
+ > start=\`date +%s\`
+ > timeout=5
+ > while [ \\( ! -f \$f \\) -a \\( ! -L \$f \\) ]; do
+ > now=\`date +%s\`
+ > if [ "\`expr \$now - \$start\`" -gt \$timeout ]; then
+ > echo "timeout: \$f was not created in \$timeout seconds (it is now \$(date +%s))"
+ > exit 1
+ > fi
+ > sleep 0.1
+ > done
+ > if [ \$# -gt 1 ]; then
+ > cat "\$@"
+ > fi
+ > EOF
+ $ chmod +x "$TESTTMP/waitlock_editor.sh"
+
+Things behave differently if we don't already have a 00changelog.i file when
+this all starts, so let's make one.
+
+ $ echo r0 > r0
+ $ hg commit -qAm 'r0'
+
+Start an hg commit that will take a while
+ $ EDITOR_STARTED="$(pwd)/.editor_started"
+ $ MISCHIEF_MANAGED="$(pwd)/.mischief_managed"
+ $ JOBS_FINISHED="$(pwd)/.jobs_finished"
+
+#if fail-if-detected
+ $ cat >> .hg/hgrc << EOF
+ > [debug]
+ > revlog.verifyposition.changelog = fail
+ > EOF
+#endif
+
+ $ echo foo > foo
+ $ (WAITLOCK_ANNOUNCE="${EDITOR_STARTED}" \
+ > WAITLOCK_FILE="${MISCHIEF_MANAGED}" \
+ > HGEDITOR="$TESTTMP/waitlock_editor.sh" \
+ > hg commit -qAm 'r1 (foo)' --edit foo > .foo_commit_out 2>&1 ; touch "${JOBS_FINISHED}") &
+
+Wait for the "editor" to actually start
+ $ WAITLOCK_FILE="${EDITOR_STARTED}" "$TESTTMP/waitlock_editor.sh"
+
+Break the locks, and make another commit.
+ $ hg debuglocks -LW
+ $ echo bar > bar
+ $ hg commit -qAm 'r2 (bar)' bar
+ $ hg debugrevlogindex -c
+ rev linkrev nodeid p1 p2
+ 0 0 222799e2f90b 000000000000 000000000000
+ 1 1 6f124f6007a0 222799e2f90b 000000000000
+
+Awaken the editor from that first commit
+ $ touch "${MISCHIEF_MANAGED}"
+And wait for it to finish
+ $ WAITLOCK_FILE="${JOBS_FINISHED}" "$TESTTMP/waitlock_editor.sh"
+
+#if skip-detection
+(Ensure there was no output)
+ $ cat .foo_commit_out
+And observe a corrupted repository -- rev 2's linkrev is 1, which should never
+happen for the changelog (the linkrev should always refer to itself).
+ $ hg debugrevlogindex -c
+ rev linkrev nodeid p1 p2
+ 0 0 222799e2f90b 000000000000 000000000000
+ 1 1 6f124f6007a0 222799e2f90b 000000000000
+ 2 1 ac80e6205bb2 222799e2f90b 000000000000
+#endif
+
+#if fail-if-detected
+ $ cat .foo_commit_out
+ transaction abort!
+ rollback completed
+ note: commit message saved in .hg/last-message.txt
+ note: use 'hg commit --logfile .hg/last-message.txt --edit' to reuse it
+ abort: 00changelog.i: file cursor at position 249, expected 121
+And no corruption in the changelog.
+ $ hg debugrevlogindex -c
+ rev linkrev nodeid p1 p2
+ 0 0 222799e2f90b 000000000000 000000000000
+ 1 1 6f124f6007a0 222799e2f90b 000000000000
+And, because of transactions, there's none in the manifestlog either.
+ $ hg debugrevlogindex -m
+ rev linkrev nodeid p1 p2
+ 0 0 7b7020262a56 000000000000 000000000000
+ 1 1 ad3fe36d86d9 7b7020262a56 000000000000
+#endif
+