view tests/test-record.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

Set up a repo

  $ cat <<EOF >> $HGRCPATH
  > [ui]
  > interactive = true
  > [extensions]
  > record =
  > EOF

  $ hg init a
  $ cd a

Record help

  $ hg record -h
  hg record [OPTION]... [FILE]...
  
  interactively select changes to commit
  
      If a list of files is omitted, all changes reported by 'hg status' will be
      candidates for recording.
  
      See 'hg help dates' for a list of formats valid for -d/--date.
  
      If using the text interface (see 'hg help config'), you will be prompted
      for whether to record changes to each modified file, and for files with
      multiple changes, for each change to use. For each query, the following
      responses are possible:
  
        y - record this change
        n - skip this change
        e - edit this change manually
  
        s - skip remaining changes to this file
        f - record remaining changes to this file
  
        d - done, skip remaining changes and files
        a - record all changes to all remaining files
        q - quit, recording no changes
  
        ? - display help
  
      This command is not available when committing a merge.
  
  (use 'hg help -e record' to show help for the record extension)
  
  options ([+] can be repeated):
  
   -A --addremove           mark new/missing files as added/removed before
                            committing
      --close-branch        mark a branch head as closed
      --amend               amend the parent of the working directory
   -s --secret              use the secret phase for committing
   -e --edit                invoke editor on commit messages
   -I --include PATTERN [+] include names matching the given patterns
   -X --exclude PATTERN [+] exclude names matching the given patterns
   -m --message TEXT        use text as commit message
   -l --logfile FILE        read commit message from file
   -d --date DATE           record the specified date as commit date
   -u --user USER           record the specified user as committer
   -S --subrepos            recurse into subrepositories
   -w --ignore-all-space    ignore white space when comparing lines
   -b --ignore-space-change ignore changes in the amount of white space
   -B --ignore-blank-lines  ignore changes whose lines are all blank
   -Z --ignore-space-at-eol ignore changes in whitespace at EOL
  
  (some details hidden, use --verbose to show complete help)

Select no files

  $ touch empty-rw
  $ hg add empty-rw

  $ hg record empty-rw<<EOF
  > n
  > EOF
  diff --git a/empty-rw b/empty-rw
  new file mode 100644
  abort: empty commit message
  [10]

  $ hg tip -p
  changeset:   -1:000000000000
  tag:         tip
  user:        
  date:        Thu Jan 01 00:00:00 1970 +0000