view tests/test-obsolete-bounds-checking.t @ 42943:5fadf6103790

remotefilelog: replace repack lock to solve race condition

2c74337e6483 reduced the probability of race-conditions when starting
background repack and prefetch and we saw the difference in our CI instance
with all failures disappearing except one where one call to waitonrepack seems
to returns too early.

I'm not sure what exactly goes wrong but I realized that while the prefetch
operation uses a standard Mercurial lock, the repack operation is using a
custom lock based on `fcntl.flock` on available platforms. As `extutil.flock`
fallback on traditional Mercurial locks on other platforms and the tests are
stable on my laptop, our CI environment and GCC112, I'm sending this patch to
standardize the behavior across environments.

Differential Revision: https://phab.mercurial-scm.org/D6844

Create a repo, set the username to something more than 255 bytes, then run hg amend on it.

  $ unset HGUSER
  $ cat >> $HGRCPATH << EOF
  > [ui]
  > username = aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa <[e-mail]>
  > [extensions]
  > amend =
  > [experimental]
  > evolution.createmarkers=True
  > evolution.exchange=True
  > EOF
  $ hg init tmpa
  $ cd tmpa
  $ echo a > a
  $ hg add
  adding a
  $ hg commit -m "Initial commit"
  $ echo a >> a
  $ hg amend 2>&1 | egrep -v '^(\*\*|  )'
  transaction abort!
  rollback completed
  Traceback (most recent call last):
  *ProgrammingError: obsstore metadata value cannot be longer than 255 bytes (value "aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa <[e-mail]>" for key "user" is 285 bytes) (glob)