Mercurial > hg
view tests/test-bheads.t @ 46607:e9901d01d135
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
author | Kyle Lippincott <spectral@google.com> |
---|---|
date | Wed, 03 Feb 2021 16:33:10 -0800 |
parents | 95c4cca641f6 |
children |
line wrap: on
line source
$ heads() > { > hg heads --template '{rev}: {desc|firstline|strip} ({branches})\n' "$@" > } $ hg init a $ cd a $ echo 'root' >root $ hg add root $ hg commit -m "Adding root node" $ heads 0: Adding root node () ------- $ heads . 0: Adding root node () ======= $ echo 'a' >a $ hg add a $ hg branch a marked working directory as branch a (branches are permanent and global, did you want a bookmark?) $ hg commit -m "Adding a branch" $ heads 1: Adding a branch (a) 0: Adding root node () ------- $ heads . 1: Adding a branch (a) ======= $ hg update -C 0 0 files updated, 0 files merged, 1 files removed, 0 files unresolved $ echo 'b' >b $ hg add b $ hg branch b marked working directory as branch b $ hg commit -m "Adding b branch" $ heads 2: Adding b branch (b) 1: Adding a branch (a) 0: Adding root node () ------- $ heads . 2: Adding b branch (b) ======= $ echo 'bh1' >bh1 $ hg add bh1 $ hg commit -m "Adding b branch head 1" $ heads 3: Adding b branch head 1 (b) 1: Adding a branch (a) 0: Adding root node () ------- $ heads . 3: Adding b branch head 1 (b) ======= $ hg update -C 2 0 files updated, 0 files merged, 1 files removed, 0 files unresolved $ echo 'bh2' >bh2 $ hg add bh2 $ hg commit -m "Adding b branch head 2" created new head $ heads 4: Adding b branch head 2 (b) 3: Adding b branch head 1 (b) 1: Adding a branch (a) 0: Adding root node () $ heads . 4: Adding b branch head 2 (b) 3: Adding b branch head 1 (b) ======= $ hg update -C 2 0 files updated, 0 files merged, 1 files removed, 0 files unresolved $ echo 'bh3' >bh3 $ hg add bh3 $ hg commit -m "Adding b branch head 3" created new head $ heads 5: Adding b branch head 3 (b) 4: Adding b branch head 2 (b) 3: Adding b branch head 1 (b) 1: Adding a branch (a) 0: Adding root node () ------- $ heads . 5: Adding b branch head 3 (b) 4: Adding b branch head 2 (b) 3: Adding b branch head 1 (b) ======= $ hg merge 4 1 files updated, 0 files merged, 0 files removed, 0 files unresolved (branch merge, don't forget to commit) $ hg commit -m "Merging b branch head 2 and b branch head 3" $ heads 6: Merging b branch head 2 and b branch head 3 (b) 3: Adding b branch head 1 (b) 1: Adding a branch (a) 0: Adding root node () ------- $ heads . 6: Merging b branch head 2 and b branch head 3 (b) 3: Adding b branch head 1 (b) ======= $ echo 'c' >c $ hg add c $ hg branch c marked working directory as branch c $ hg commit -m "Adding c branch" $ heads 7: Adding c branch (c) 6: Merging b branch head 2 and b branch head 3 (b) 3: Adding b branch head 1 (b) 1: Adding a branch (a) 0: Adding root node () ------- $ heads . 7: Adding c branch (c) ======= $ heads -r 3 . no open branch heads found on branches c (started at 3) [1] $ heads -r 2 . 7: Adding c branch (c) ------- $ hg update -C 4 0 files updated, 0 files merged, 2 files removed, 0 files unresolved ------- $ heads -r 3 . 3: Adding b branch head 1 (b) ------- $ heads -r 2 . 6: Merging b branch head 2 and b branch head 3 (b) 3: Adding b branch head 1 (b) ------- $ heads -r 7 . no open branch heads found on branches b (started at 7) [1] ======= $ for i in 0 1 2 3 4 5 6 7; do > hg update -C "$i" > heads > echo '-------' > heads . > echo '-------' > done 0 files updated, 0 files merged, 2 files removed, 0 files unresolved 7: Adding c branch (c) 6: Merging b branch head 2 and b branch head 3 (b) 3: Adding b branch head 1 (b) 1: Adding a branch (a) 0: Adding root node () ------- 0: Adding root node () ------- 1 files updated, 0 files merged, 0 files removed, 0 files unresolved 7: Adding c branch (c) 6: Merging b branch head 2 and b branch head 3 (b) 3: Adding b branch head 1 (b) 1: Adding a branch (a) 0: Adding root node () ------- 1: Adding a branch (a) ------- 1 files updated, 0 files merged, 1 files removed, 0 files unresolved 7: Adding c branch (c) 6: Merging b branch head 2 and b branch head 3 (b) 3: Adding b branch head 1 (b) 1: Adding a branch (a) 0: Adding root node () ------- 6: Merging b branch head 2 and b branch head 3 (b) 3: Adding b branch head 1 (b) ------- 1 files updated, 0 files merged, 0 files removed, 0 files unresolved 7: Adding c branch (c) 6: Merging b branch head 2 and b branch head 3 (b) 3: Adding b branch head 1 (b) 1: Adding a branch (a) 0: Adding root node () ------- 6: Merging b branch head 2 and b branch head 3 (b) 3: Adding b branch head 1 (b) ------- 1 files updated, 0 files merged, 1 files removed, 0 files unresolved 7: Adding c branch (c) 6: Merging b branch head 2 and b branch head 3 (b) 3: Adding b branch head 1 (b) 1: Adding a branch (a) 0: Adding root node () ------- 6: Merging b branch head 2 and b branch head 3 (b) 3: Adding b branch head 1 (b) ------- 1 files updated, 0 files merged, 1 files removed, 0 files unresolved 7: Adding c branch (c) 6: Merging b branch head 2 and b branch head 3 (b) 3: Adding b branch head 1 (b) 1: Adding a branch (a) 0: Adding root node () ------- 6: Merging b branch head 2 and b branch head 3 (b) 3: Adding b branch head 1 (b) ------- 1 files updated, 0 files merged, 0 files removed, 0 files unresolved 7: Adding c branch (c) 6: Merging b branch head 2 and b branch head 3 (b) 3: Adding b branch head 1 (b) 1: Adding a branch (a) 0: Adding root node () ------- 6: Merging b branch head 2 and b branch head 3 (b) 3: Adding b branch head 1 (b) ------- 1 files updated, 0 files merged, 0 files removed, 0 files unresolved 7: Adding c branch (c) 6: Merging b branch head 2 and b branch head 3 (b) 3: Adding b branch head 1 (b) 1: Adding a branch (a) 0: Adding root node () ------- 7: Adding c branch (c) ------- ======= $ for i in a b c z; do > heads "$i" > echo '-------' > done 1: Adding a branch (a) ------- 6: Merging b branch head 2 and b branch head 3 (b) 3: Adding b branch head 1 (b) ------- 7: Adding c branch (c) ------- abort: unknown revision 'z' ------- ======= $ heads 0 1 2 3 4 5 6 7 7: Adding c branch (c) 6: Merging b branch head 2 and b branch head 3 (b) 3: Adding b branch head 1 (b) 1: Adding a branch (a) 0: Adding root node () Topological heads: $ heads -t 7: Adding c branch (c) 3: Adding b branch head 1 (b) 1: Adding a branch (a) $ cd .. ______________ "created new head" message tests $ hg init newheadmsg $ cd newheadmsg Init: no msg $ echo 1 > a $ hg ci -Am "a0: Initial root" adding a $ echo 2 >> a $ hg ci -m "a1 (HN)" $ hg branch b marked working directory as branch b (branches are permanent and global, did you want a bookmark?) $ echo 1 > b $ hg ci -Am "b2: Initial root for branch b" adding b $ echo 2 >> b $ hg ci -m "b3 (HN)" Case NN: msg $ hg up -q null $ hg branch -f b marked working directory as branch b $ echo 1 > bb $ hg ci -Am "b4 (NN): new topo root for branch b" adding bb created new head Case HN: no msg $ echo 2 >> bb $ hg ci -m "b5 (HN)" Case BN: msg $ hg branch -f default marked working directory as branch default $ echo 1 > aa $ hg ci -Am "a6 (BN): new branch root" adding aa created new head Case CN: msg $ hg up -q 4 $ echo 3 >> bbb $ hg ci -Am "b7 (CN): regular new head" adding bbb created new head Case BB: msg $ hg up -q 4 $ hg merge -q 3 $ hg branch -f default marked working directory as branch default $ hg ci -m "a8 (BB): weird new branch root" created new head Case CB: msg $ hg up -q 4 $ hg merge -q 1 $ hg ci -m "b9 (CB): new head from branch merge" created new head Case HB: no msg $ hg up -q 7 $ hg merge -q 6 $ hg ci -m "b10 (HB): continuing head from branch merge" Case CC: msg $ hg up -q 4 $ hg merge -q 2 $ hg ci -m "b11 (CC): new head from merge" created new head Case CH: no msg $ hg up -q 2 $ hg merge -q 10 $ hg ci -m "b12 (CH): continuing head from merge" Case HH: no msg $ hg merge -q 3 $ hg ci -m "b12 (HH): merging two heads" $ cd ..