diff options
author | SZEDER Gábor <szeder.dev@gmail.com> | 2018-08-01 01:32:48 +0200 |
---|---|---|
committer | Junio C Hamano <gitster@pobox.com> | 2018-08-01 10:07:21 -0700 |
commit | 377d845943c989543b0c6673e2bcd6ba2de83b27 (patch) | |
tree | 73db7484e7371e82689426d88bc2f288e4f242cd /Documentation/RelNotes/1.9.2.txt | |
parent | Fourth batch for 2.19 cycle (diff) | |
download | tgif-377d845943c989543b0c6673e2bcd6ba2de83b27.tar.xz |
t1404: increase core.packedRefsTimeout to avoid occasional test failure
The test 'no bogus intermediate values during delete' in
't1404-update-ref-errors.sh', added in 6a2a7736d8 (t1404: demonstrate
two problems with reference transactions, 2017-09-08), tries to catch
undesirable side effects of deleting a ref, both loose and packed, in
a transaction. To do so it is holding the packed refs file locked
when it starts 'git update-ref -d' in the background with a 3secs
'core.packedRefsTimeout' value. After performing a few checks it is
then supposed to unlock the packed refs file before the background
'git update-ref's attempt to acquire the lock times out.
While 3secs timeout seems plenty, and indeed is sufficient in most
cases, on rare occasions it's just not quite enough: I saw this test
fail in Travis CI build jobs two, maybe three times because 'git
update-ref' timed out.
Increase that timeout by an order of magnitude to 30s to make such an
occasional failure even more improbable. This won't make the test run
any longer under normal circumstances, because 'git update-ref' will
acquire the lock and resume execution as soon as it can. And if it
turns out that even this increased timeout is still not enough, then
there are most likely bigger problems, e.g. the Travis CI build job
will exceed its time limit anyway, or the lockfile module is broken.
Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Diffstat (limited to 'Documentation/RelNotes/1.9.2.txt')
0 files changed, 0 insertions, 0 deletions