GNU bug report logs - #76187
vc-git-test-dir-branch-headers failure on Fedora

Previous Next

Full log

Message #35 received at 76187 <at> debbugs.gnu.org (full text, mbox):

From: Kévin Le Gouguec <kevin.legouguec <at> gmail.com>
To: Paul Eggert <eggert <at> cs.ucla.edu>
Cc: 76187 <at> debbugs.gnu.org, Eli Zaretskii <eliz <at> gnu.org>
Subject: Re: bug#76187: vc-git-test-dir-branch-headers failure on Fedora
Date: Wed, 12 Feb 2025 23:37:19 +0100

[Message part 1 (text/plain, inline)]

Kévin Le Gouguec <kevin.legouguec <at> gmail.com> writes:

>> My current working theories are:
>>
>> a. We fail to sync with some asynchronous processes spawned by vc before
>>    running these 'git checkout' commands; there is precedent for
>>    synchronization problems:
>>
[…]
>>
>> b. One of the test's git commands spawns a… background GC job or
>>    something, and the test's next git command trips over that.
>
> c. That vc-dir-kill-dir-status-process actually kills a Git process that
>    is "holding the lock", in which case…
>
>> I suppose either a or b would be solved by vc-git-test--run spinning for
>> a bit while (file-exists-p "…/index.lock"); ideally though I'd prefer to
>> catch the locking process red-handed… not sure how to achieve this
>> though.
>
> … this will not work.
>
> I'd like to believe that the present bug stems from that dodgy FIXME
> workaround, and addressing the latter will fix the former, but that
> FIXME was an admission of defeat already.  Haven't grown wiser since
> then unfortunately.

Bafflement continues.  With the attached patch that

* removes the call to vc-dir-kill-dir-status-process to eliminate
hypothesis c,
* replaces it with a (while (vc-dir-busy) …) loop to mitigate a,
* adds a (while (file-exists-p ".git/index.lock")) loop to mitigate b,

I still reproduce.  Here, after 700 runs in 6min:

      GEN      lisp/vc/vc-git-tests.log
    Running 8 tests (2025-02-12 23:16:08+0100, selector `(not (or (tag :unstable) (tag :nativecomp)))')
       passed  1/8  vc-git-test-annotate-time (0.002284 sec)
    Paused 1 time(s) waiting for vc-dir-busy
    err in /tmp/emacs-test-dhul2y-vc-git/: (error Failed (status 128): git --no-pager checkout -b feature/foo master .)
    (buffer-string:
    fatal: Unable to create '/tmp/emacs-test-dhul2y-vc-git/.git/index.lock': File exists.

    Another git process seems to be running in this repository, e.g.
    an editor opened by 'git commit'. Please make sure all processes
    are terminated then try again. If it still fails, a git process
    may have crashed in this repository earlier:
    remove the file manually to continue.

    )

I have a hard time making sense of the sequence of events.  Evidently:

1. vc-git-test--run did *not* pause for (file-exists-p
   ".git/index.lock"),
2. but Git *still* tripped over a preexisting .git/index.lock.

Logging default-directory in the condition-case error handler confirms
we are in the correct directory, so I do not understand why
file-exists-p did not see the lock that Git reports.  All I can think of
is a stray Git process ¿started at some point by someone? that grabs the
lock between steps 1 & 2.

Barring "turn off background processing" knobs for VC and/or Git that I
missed, not sure what to try next beside catching that "index.lock
exists" error and re-trying after a delay.

[debug.patch (text/x-patch, attachment)]

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.