* [CI]: Is t7527 known to be flakey?
@ 2023-01-20 2:52 Junio C Hamano
2023-01-20 15:23 ` Jeff Hostetler
2023-01-21 10:23 ` SZEDER Gábor
0 siblings, 2 replies; 8+ messages in thread
From: Junio C Hamano @ 2023-01-20 2:52 UTC (permalink / raw)
To: Jeff Hostetler; +Cc: git
The said test failed its linux-musl job in its first attempt, but
re-running the failed job passed.
https://github.com/git/git/actions/runs/3963948890/jobs/6792356234
(seen@e096683 attempt #1 linux-musl)
https://github.com/git/git/actions/runs/3963948890/jobs/6792850313
(seen@e096683 attempt #2 linux-musl)
^ permalink raw reply [flat|nested] 8+ messages in thread* Re: [CI]: Is t7527 known to be flakey? 2023-01-20 2:52 [CI]: Is t7527 known to be flakey? Junio C Hamano @ 2023-01-20 15:23 ` Jeff Hostetler 2023-01-20 15:40 ` Junio C Hamano 2023-01-21 10:23 ` SZEDER Gábor 1 sibling, 1 reply; 8+ messages in thread From: Jeff Hostetler @ 2023-01-20 15:23 UTC (permalink / raw) To: Junio C Hamano, Jeff Hostetler, edecosta; +Cc: git On 1/19/23 9:52 PM, Junio C Hamano wrote: > The said test failed its linux-musl job in its first attempt, but > re-running the failed job passed. > > https://github.com/git/git/actions/runs/3963948890/jobs/6792356234 > (seen@e096683 attempt #1 linux-musl) > > https://github.com/git/git/actions/runs/3963948890/jobs/6792850313 > (seen@e096683 attempt #2 linux-musl) > This is on Linux, so it would be using the linux inotify backend. Let me add Eric to the "To:" line for visibility. And see if he has experienced this during his development of it. I've not looked at the inotify code so I can't say if there are races there or not. Tests that move directories feel like good candidates for race conditions -- since the daemon doesn't get a recursive view of the tree with inotify() and must simulate that and manage the individual directories, but again I don't want to assume that. Jeff ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [CI]: Is t7527 known to be flakey? 2023-01-20 15:23 ` Jeff Hostetler @ 2023-01-20 15:40 ` Junio C Hamano 0 siblings, 0 replies; 8+ messages in thread From: Junio C Hamano @ 2023-01-20 15:40 UTC (permalink / raw) To: Jeff Hostetler; +Cc: Jeff Hostetler, edecosta, git Jeff Hostetler <git@jeffhostetler.com> writes: > This is on Linux, so it would be using the linux inotify backend. > Let me add Eric to the "To:" line for visibility. Thanks for redirection. ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [CI]: Is t7527 known to be flakey? 2023-01-20 2:52 [CI]: Is t7527 known to be flakey? Junio C Hamano 2023-01-20 15:23 ` Jeff Hostetler @ 2023-01-21 10:23 ` SZEDER Gábor 2023-01-23 16:56 ` Jeff Hostetler 1 sibling, 1 reply; 8+ messages in thread From: SZEDER Gábor @ 2023-01-21 10:23 UTC (permalink / raw) To: Junio C Hamano; +Cc: Jeff Hostetler, edecosta, git On Thu, Jan 19, 2023 at 06:52:01PM -0800, Junio C Hamano wrote: > The said test failed its linux-musl job in its first attempt, but > re-running the failed job passed. > > https://github.com/git/git/actions/runs/3963948890/jobs/6792356234 > (seen@e096683 attempt #1 linux-musl) > > https://github.com/git/git/actions/runs/3963948890/jobs/6792850313 > (seen@e096683 attempt #2 linux-musl) t7527 is quite slow, even with the right selection of test cases, but this little tweak makes it much faster: diff --git a/t/t7527-builtin-fsmonitor.sh b/t/t7527-builtin-fsmonitor.sh index 0e497ba98d..4210ef644c 100755 --- a/t/t7527-builtin-fsmonitor.sh +++ b/t/t7527-builtin-fsmonitor.sh @@ -676,7 +676,10 @@ test_expect_success 'cleanup worktrees' ' # cause incorrect results when the untracked-cache is enabled. test_lazy_prereq UNTRACKED_CACHE ' - git update-index --test-untracked-cache + # This check takes a very long time, but I know it works on + # my system, so let's fake it. + #git update-index --test-untracked-cache + true ' test_expect_success 'Matrix: setup for untracked-cache,fsmonitor matrix' ' This with the right selection of test cases makes --stress practicable, and the test tends to fail after a handful of repetitions: ./t7527-builtin-fsmonitor.sh --stress -r 8,23-58 I saw different failures in multiple test cases, e.g.: Unexpected output in case 55: expecting success of 7527.55 'Matrix[uc:true][fsm:true] move_directory_contents_deeper': matrix_clean_up_repo && $fn && if test $uc = false && test $fsm = false then git status --porcelain=v1 >.git/expect.$fn else git status --porcelain=v1 >.git/actual.$fn && test_cmp .git/expect.$fn .git/actual.$fn fi + matrix_clean_up_repo + git reset --hard HEAD HEAD is now at 1d1edcb initial + git clean -fd + move_directory_contents_deeper + mkdir T1/_new_ + mv T1/F1 T1/F2 T1/T2 T1/_new_ + test true = false + git status --porcelain=v1 error: read error: Connection reset by peer error: could not read IPC response + test_cmp .git/expect.move_directory_contents_deeper .git/actual.move_directory_contents_deeper + test 2 -ne 2 + eval diff -u "$@" + diff -u .git/expect.move_directory_contents_deeper .git/actual.move_directory_contents_deeper --- .git/expect.move_directory_contents_deeper 2023-01-21 09:47:12.677410349 +0000 +++ .git/actual.move_directory_contents_deeper 2023-01-21 09:47:14.045448573 +0000 @@ -7,3 +7,4 @@ D T1/T2/T3/T4/F1 D T1/T2/T3/T4/F2 ?? T1/_new_/ +?? dir1 error: last command exited with $?=1 not ok 55 - Matrix[uc:true][fsm:true] move_directory_contents_deeper SIGPIPE in 'git status' cases 42, 43 and 55: expecting success of 7527.42 'Matrix[uc:false][fsm:true] move_directory_contents_deeper': matrix_clean_up_repo && $fn && if test $uc = false && test $fsm = false then git status --porcelain=v1 >.git/expect.$fn else git status --porcelain=v1 >.git/actual.$fn && test_cmp .git/expect.$fn .git/actual.$fn fi + matrix_clean_up_repo + git reset --hard HEAD HEAD is now at 1d1edcb initial + git clean -fd + move_directory_contents_deeper + mkdir T1/_new_ + mv T1/F1 T1/F2 T1/T2 T1/_new_ + test false = false + test true = false + git status --porcelain=v1 error: last command exited with $?=141 not ok 42 - Matrix[uc:false][fsm:true] move_directory_contents_deeper expecting success of 7527.43 'Matrix[uc:false][fsm:true] move_directory_up': matrix_clean_up_repo && $fn && if test $uc = false && test $fsm = false then git status --porcelain=v1 >.git/expect.$fn else git status --porcelain=v1 >.git/actual.$fn && test_cmp .git/expect.$fn .git/actual.$fn fi + matrix_clean_up_repo + git reset --hard HEAD HEAD is now at 1d1edcb initial + git clean -fd Removing T1/_new_/ + move_directory_up + mv T1/T2/T3 T1 + test false = false + test true = false + git status --porcelain=v1 error: last command exited with $?=141 not ok 43 - Matrix[uc:false][fsm:true] move_directory_up expecting success of 7527.55 'Matrix[uc:true][fsm:true] move_directory_contents_deeper': matrix_clean_up_repo && $fn && if test $uc = false && test $fsm = false then git status --porcelain=v1 >.git/expect.$fn else git status --porcelain=v1 >.git/actual.$fn && test_cmp .git/expect.$fn .git/actual.$fn fi + matrix_clean_up_repo + git reset --hard HEAD HEAD is now at 1d1edcb initial + git clean -fd + move_directory_contents_deeper + mkdir T1/_new_ + mv T1/F1 T1/F2 T1/T2 T1/_new_ + test true = false + git status --porcelain=v1 error: last command exited with $?=141 I find it interesting that the output of 'git status' is redirected to a file in all these cases, and yet it gets a SIGPIPE. I also saw the test hang a couple of times, e.g.: $ ./t7527-builtin-fsmonitor.sh --stress -r 8,23-58 OK 6.1 OK 7.1 OK 1.1 OK 2.1 OK 3.1 OK 5.1 OK 4.1 OK 0.1 OK 6.2 OK 1.2 OK 2.2 OK 7.2 OK 5.2 OK 0.2 OK 4.2 OK 6.3 OK 7.3 OK 2.3 OK 0.3 OK 4.3 OK 6.4 OK 7.4 OK 2.4 OK 0.4 OK 4.4 OK 6.5 OK 7.5 OK 2.5 OK 0.5 OK 4.5 OK 6.6 OK 7.6 OK 2.6 OK 0.6 OK 4.6 OK 6.7 OK 7.7 OK 2.7 OK 0.7 OK 4.7 OK 6.8 OK 7.8 OK 2.8 OK 0.8 OK 4.8 OK 6.9 OK 7.9 OK 2.9 OK 0.9 OK 4.9 OK 6.10 OK 7.10 OK 2.10 OK 0.10 OK 4.10 OK 6.11 OK 7.11 OK 2.11 OK 0.11 OK 4.11 OK 6.12 OK 7.12 OK 2.12 OK 0.12 OK 4.12 OK 6.13 OK 7.13 OK 2.13 OK 0.13 OK 6.14 OK 4.13 OK 7.14 OK 2.14 OK 0.14 OK 6.15 OK 4.14 OK 2.15 OK 7.15 OK 0.15 FAIL 7.16 OK 6.16 OK 2.16 OK 4.15 OK 0.16 At this point the test script should print the log of the failed job, but it hangs instead, as there are a number of stuck fsmonitor--daemon and status processes (notice how the stress test starts with 8 jobs, but the last repetition only has 4): $ ps aux |grep git szeder 1857100 0.0 0.1 72272 4452 pts/2 Sl+ 21:40 0:00 /home/szeder/src/git/git fsmonitor--daemon run --detach --ipc-threads=8 szeder 1857779 0.0 0.1 6560 4152 pts/2 S+ 21:40 0:00 /home/szeder/src/git/git status --porcelain=v1 szeder 1860020 0.0 0.1 88664 4312 pts/2 Sl+ 21:40 0:00 /home/szeder/src/git/git fsmonitor--daemon run --detach --ipc-threads=8 szeder 1860668 0.0 0.1 6560 4040 pts/2 S+ 21:40 0:00 /home/szeder/src/git/git status --porcelain=v1 szeder 1860749 0.0 0.1 96860 4528 pts/2 Sl+ 21:40 0:00 /home/szeder/src/git/git fsmonitor--daemon run --detach --ipc-threads=8 szeder 1861281 0.0 0.1 6560 4272 pts/2 S+ 21:40 0:00 /home/szeder/src/git/git status --porcelain=v1 ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [CI]: Is t7527 known to be flakey? 2023-01-21 10:23 ` SZEDER Gábor @ 2023-01-23 16:56 ` Jeff Hostetler 2023-01-23 18:12 ` SZEDER Gábor 0 siblings, 1 reply; 8+ messages in thread From: Jeff Hostetler @ 2023-01-23 16:56 UTC (permalink / raw) To: SZEDER Gábor, Junio C Hamano; +Cc: Jeff Hostetler, edecosta, git On 1/21/23 5:23 AM, SZEDER Gábor wrote: > On Thu, Jan 19, 2023 at 06:52:01PM -0800, Junio C Hamano wrote: >> The said test failed its linux-musl job in its first attempt, but >> re-running the failed job passed. >> >> https://github.com/git/git/actions/runs/3963948890/jobs/6792356234 >> (seen@e096683 attempt #1 linux-musl) >> >> https://github.com/git/git/actions/runs/3963948890/jobs/6792850313 >> (seen@e096683 attempt #2 linux-musl) > > t7527 is quite slow, even with the right selection of test cases, but > this little tweak makes it much faster: > > diff --git a/t/t7527-builtin-fsmonitor.sh b/t/t7527-builtin-fsmonitor.sh > index 0e497ba98d..4210ef644c 100755 > --- a/t/t7527-builtin-fsmonitor.sh > +++ b/t/t7527-builtin-fsmonitor.sh > @@ -676,7 +676,10 @@ test_expect_success 'cleanup worktrees' ' > # cause incorrect results when the untracked-cache is enabled. > > test_lazy_prereq UNTRACKED_CACHE ' > - git update-index --test-untracked-cache > + # This check takes a very long time, but I know it works on > + # my system, so let's fake it. > + #git update-index --test-untracked-cache > + true > ' > > test_expect_success 'Matrix: setup for untracked-cache,fsmonitor matrix' ' > > This with the right selection of test cases makes --stress > practicable, and the test tends to fail after a handful of > repetitions: > > ./t7527-builtin-fsmonitor.sh --stress -r 8,23-58 > > I saw different failures in multiple test cases, e.g.: > > Unexpected output in case 55: > > expecting success of 7527.55 'Matrix[uc:true][fsm:true] move_directory_contents_deeper': > matrix_clean_up_repo && > $fn && > if test $uc = false && test $fsm = false > then > git status --porcelain=v1 >.git/expect.$fn > else > git status --porcelain=v1 >.git/actual.$fn && > test_cmp .git/expect.$fn .git/actual.$fn > fi > > + matrix_clean_up_repo > + git reset --hard HEAD > HEAD is now at 1d1edcb initial > + git clean -fd > + move_directory_contents_deeper > + mkdir T1/_new_ > + mv T1/F1 T1/F2 T1/T2 T1/_new_ > + test true = false > + git status --porcelain=v1 > error: read error: Connection reset by peer > error: could not read IPC response > + test_cmp .git/expect.move_directory_contents_deeper .git/actual.move_directory_contents_deeper > + test 2 -ne 2 > + eval diff -u "$@" > + diff -u .git/expect.move_directory_contents_deeper .git/actual.move_directory_contents_deeper > --- .git/expect.move_directory_contents_deeper 2023-01-21 09:47:12.677410349 +0000 > +++ .git/actual.move_directory_contents_deeper 2023-01-21 09:47:14.045448573 +0000 > @@ -7,3 +7,4 @@ > D T1/T2/T3/T4/F1 > D T1/T2/T3/T4/F2 > ?? T1/_new_/ > +?? dir1 > error: last command exited with $?=1 > not ok 55 - Matrix[uc:true][fsm:true] move_directory_contents_deeper > > SIGPIPE in 'git status' cases 42, 43 and 55: > > expecting success of 7527.42 'Matrix[uc:false][fsm:true] move_directory_contents_deeper': > matrix_clean_up_repo && > $fn && > if test $uc = false && test $fsm = false > then > git status --porcelain=v1 >.git/expect.$fn > else > git status --porcelain=v1 >.git/actual.$fn && > test_cmp .git/expect.$fn .git/actual.$fn > fi > > + matrix_clean_up_repo > + git reset --hard HEAD > HEAD is now at 1d1edcb initial > + git clean -fd > + move_directory_contents_deeper > + mkdir T1/_new_ > + mv T1/F1 T1/F2 T1/T2 T1/_new_ > + test false = false > + test true = false > + git status --porcelain=v1 > error: last command exited with $?=141 > not ok 42 - Matrix[uc:false][fsm:true] move_directory_contents_deeper > > expecting success of 7527.43 'Matrix[uc:false][fsm:true] move_directory_up': > matrix_clean_up_repo && > $fn && > if test $uc = false && test $fsm = false > then > git status --porcelain=v1 >.git/expect.$fn > else > git status --porcelain=v1 >.git/actual.$fn && > test_cmp .git/expect.$fn .git/actual.$fn > fi > > + matrix_clean_up_repo > + git reset --hard HEAD > HEAD is now at 1d1edcb initial > + git clean -fd > Removing T1/_new_/ > + move_directory_up > + mv T1/T2/T3 T1 > + test false = false > + test true = false > + git status --porcelain=v1 > error: last command exited with $?=141 > not ok 43 - Matrix[uc:false][fsm:true] move_directory_up > > expecting success of 7527.55 'Matrix[uc:true][fsm:true] move_directory_contents_deeper': > matrix_clean_up_repo && > $fn && > if test $uc = false && test $fsm = false > then > git status --porcelain=v1 >.git/expect.$fn > else > git status --porcelain=v1 >.git/actual.$fn && > test_cmp .git/expect.$fn .git/actual.$fn > fi > > + matrix_clean_up_repo > + git reset --hard HEAD > HEAD is now at 1d1edcb initial > + git clean -fd > + move_directory_contents_deeper > + mkdir T1/_new_ > + mv T1/F1 T1/F2 T1/T2 T1/_new_ > + test true = false > + git status --porcelain=v1 > error: last command exited with $?=141 > > I find it interesting that the output of 'git status' is redirected to > a file in all these cases, and yet it gets a SIGPIPE. The `git status` command talks to the running `git fsmonitor--daemon` on a Unix domain socket (or a Named Pipe on Windows), so a SIGPIPE is possible. Was this on Linux or MacOS ? Jeff > > > I also saw the test hang a couple of times, e.g.: > > $ ./t7527-builtin-fsmonitor.sh --stress -r 8,23-58 > OK 6.1 > OK 7.1 > OK 1.1 > OK 2.1 > OK 3.1 > OK 5.1 > OK 4.1 > OK 0.1 > OK 6.2 > OK 1.2 > OK 2.2 > OK 7.2 > OK 5.2 > OK 0.2 > OK 4.2 > OK 6.3 > OK 7.3 > OK 2.3 > OK 0.3 > OK 4.3 > OK 6.4 > OK 7.4 > OK 2.4 > OK 0.4 > OK 4.4 > OK 6.5 > OK 7.5 > OK 2.5 > OK 0.5 > OK 4.5 > OK 6.6 > OK 7.6 > OK 2.6 > OK 0.6 > OK 4.6 > OK 6.7 > OK 7.7 > OK 2.7 > OK 0.7 > OK 4.7 > OK 6.8 > OK 7.8 > OK 2.8 > OK 0.8 > OK 4.8 > OK 6.9 > OK 7.9 > OK 2.9 > OK 0.9 > OK 4.9 > OK 6.10 > OK 7.10 > OK 2.10 > OK 0.10 > OK 4.10 > OK 6.11 > OK 7.11 > OK 2.11 > OK 0.11 > OK 4.11 > OK 6.12 > OK 7.12 > OK 2.12 > OK 0.12 > OK 4.12 > OK 6.13 > OK 7.13 > OK 2.13 > OK 0.13 > OK 6.14 > OK 4.13 > OK 7.14 > OK 2.14 > OK 0.14 > OK 6.15 > OK 4.14 > OK 2.15 > OK 7.15 > OK 0.15 > FAIL 7.16 > OK 6.16 > OK 2.16 > OK 4.15 > OK 0.16 > > At this point the test script should print the log of the failed job, > but it hangs instead, as there are a number of stuck fsmonitor--daemon > and status processes (notice how the stress test starts with 8 jobs, > but the last repetition only has 4): > > $ ps aux |grep git > szeder 1857100 0.0 0.1 72272 4452 pts/2 Sl+ 21:40 0:00 /home/szeder/src/git/git fsmonitor--daemon run --detach --ipc-threads=8 > szeder 1857779 0.0 0.1 6560 4152 pts/2 S+ 21:40 0:00 /home/szeder/src/git/git status --porcelain=v1 > szeder 1860020 0.0 0.1 88664 4312 pts/2 Sl+ 21:40 0:00 /home/szeder/src/git/git fsmonitor--daemon run --detach --ipc-threads=8 > szeder 1860668 0.0 0.1 6560 4040 pts/2 S+ 21:40 0:00 /home/szeder/src/git/git status --porcelain=v1 > szeder 1860749 0.0 0.1 96860 4528 pts/2 Sl+ 21:40 0:00 /home/szeder/src/git/git fsmonitor--daemon run --detach --ipc-threads=8 > szeder 1861281 0.0 0.1 6560 4272 pts/2 S+ 21:40 0:00 /home/szeder/src/git/git status --porcelain=v1 > ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [CI]: Is t7527 known to be flakey? 2023-01-23 16:56 ` Jeff Hostetler @ 2023-01-23 18:12 ` SZEDER Gábor 2023-01-25 19:02 ` Jeff Hostetler 0 siblings, 1 reply; 8+ messages in thread From: SZEDER Gábor @ 2023-01-23 18:12 UTC (permalink / raw) To: Jeff Hostetler; +Cc: Junio C Hamano, Jeff Hostetler, edecosta, git On Mon, Jan 23, 2023 at 11:56:53AM -0500, Jeff Hostetler wrote: > Was this on Linux or MacOS ? On an average-ish Linux (an Ubuntu LTS variant). So the issue is not specific to musl. ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [CI]: Is t7527 known to be flakey? 2023-01-23 18:12 ` SZEDER Gábor @ 2023-01-25 19:02 ` Jeff Hostetler 2023-01-25 21:12 ` SZEDER Gábor 0 siblings, 1 reply; 8+ messages in thread From: Jeff Hostetler @ 2023-01-25 19:02 UTC (permalink / raw) To: SZEDER Gábor; +Cc: Junio C Hamano, Jeff Hostetler, edecosta, git On 1/23/23 1:12 PM, SZEDER Gábor wrote: > On Mon, Jan 23, 2023 at 11:56:53AM -0500, Jeff Hostetler wrote: >> Was this on Linux or MacOS ? > > On an average-ish Linux (an Ubuntu LTS variant). So the issue is not > specific to musl. > OK, thanks. I wasn't worried about "musl", but rather whether you were running the stress test on a Linux or Mac. Since they have different backends (inotify vs FSEvent) and all the code that touches the filesystem is different, it would best to start on the correct OS when trying to repro it. Can you tell from your stess test whether the fsmonitor-daemon is crashing? (It might be subtle since the daemon is auto-started if necessary, so it might be crashing and silently getting restarted by the next command.) I ask because a SIGPIPE in the client would make me think that the server suddenly closed the connection unexpectedly, like if it had SIGSEGV'd or something. I won't have time to spin up a Linux VM until next week, so I won't be able to investigate this for a bit. Jeff ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [CI]: Is t7527 known to be flakey? 2023-01-25 19:02 ` Jeff Hostetler @ 2023-01-25 21:12 ` SZEDER Gábor 0 siblings, 0 replies; 8+ messages in thread From: SZEDER Gábor @ 2023-01-25 21:12 UTC (permalink / raw) To: Jeff Hostetler; +Cc: Junio C Hamano, Jeff Hostetler, edecosta, git On Wed, Jan 25, 2023 at 02:02:40PM -0500, Jeff Hostetler wrote: > Can you tell from your stess test whether the fsmonitor-daemon > is crashing? (It might be subtle since the daemon is auto-started > if necessary, so it might be crashing and silently getting restarted > by the next command.) > > I ask because a SIGPIPE in the client would make me think that the > server suddenly closed the connection unexpectedly, like if it had > SIGSEGV'd or something. Last time around I only looked at the failing test case, and didn't notice anything that might have indicated the cause of the SIGPIPE. This time I chanced to look a bit further up in the test log, and: expecting success of 7527.55 'Matrix[uc:true][fsm:true] move_directory_contents_deeper': matrix_clean_up_repo && $fn && if test $uc = false && test $fsm = false then git status --porcelain=v1 >.git/expect.$fn else git status --porcelain=v1 >.git/actual.$fn && test_cmp .git/expect.$fn .git/actual.$fn fi + matrix_clean_up_repo + git reset --hard HEAD HEAD is now at 1d1edcb initial + git clean -fd + move_directory_contents_deeper + mkdir T1/_new_ + mv T1/F1 T1/F2 T1/T2 T1/_new_ + test true = false + git status --porcelain=v1 error: read error: Connection reset by peer error: could not read IPC response + test_cmp .git/expect.move_directory_contents_deeper .git/actual.move_directory_contents_deeper + test 2 -ne 2 + eval diff -u "$@" + diff -u .git/expect.move_directory_contents_deeper .git/actual.move_directory_contents_deeper ok 55 - Matrix[uc:true][fsm:true] move_directory_contents_deeper expecting success of 7527.56 'Matrix[uc:true][fsm:true] move_directory_up': matrix_clean_up_repo && $fn && if test $uc = false && test $fsm = false then git status --porcelain=v1 >.git/expect.$fn else git status --porcelain=v1 >.git/actual.$fn && test_cmp .git/expect.$fn .git/actual.$fn fi + matrix_clean_up_repo + git reset --hard HEAD HEAD is now at 1d1edcb initial + git clean -fd Removing T1/_new_/ + move_directory_up + mv T1/T2/T3 T1 + test true = false + git status --porcelain=v1 error: last command exited with $?=141 not ok 56 - Matrix[uc:true][fsm:true] move_directory_up Notice that "error: read error: Connection reset by peer" in the previous, still successful test case! I ran it a couple of times, and saw the same error message in the still successful '42 - Matrix[uc:false][fsm:true] move_directory_contents_deeper' followed by a SIGPIPE caused failure in the next test case. And now that I knew what to look for, I noticed this error message in the very first test failure I reported the other day, which didn't fail because of SIGPIPE, and in that case the error message was printed in the failed test case. And there were a few cases that failed because of SIGPIPE but there were no error messages at all. I can't say what caused these errors, but I doubt that anything segfaulted, because segfaults are logged in syslog, and I haven't found any such syslog entries coinciding with stress testing. ^ permalink raw reply [flat|nested] 8+ messages in thread
end of thread, other threads:[~2023-01-25 21:13 UTC | newest] Thread overview: 8+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2023-01-20 2:52 [CI]: Is t7527 known to be flakey? Junio C Hamano 2023-01-20 15:23 ` Jeff Hostetler 2023-01-20 15:40 ` Junio C Hamano 2023-01-21 10:23 ` SZEDER Gábor 2023-01-23 16:56 ` Jeff Hostetler 2023-01-23 18:12 ` SZEDER Gábor 2023-01-25 19:02 ` Jeff Hostetler 2023-01-25 21:12 ` SZEDER Gábor
Code repositories for project(s) associated with this public inbox https://80x24.org/mirrors/git.git This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).