go: runtime: frequent TestCrashDumpsAllThreads failures on `linux/s390x`

#!watchflakes
post <- goarch == "s390x" && pkg == "runtime" && test == "TestCrashDumpsAllThreads"

Go version

ca804118b1ff4ea6cd0b83808e140671d7e10f6f

What operating system and processor architecture are you using (`go env`)?

linux-s390x-ibm-race at ca804118b1ff4ea6cd0b83808e140671d7e10f6f

:: Running /data/golang/workdir/go/src/make.bash with args ["/data/golang/workdir/go/src/make.bash" "-force"] and env ["LANG=en_US.UTF-8" "PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin" "HOME=/root" "LOGNAME=root" "USER=root" "SHELL=/bin/sh" "INVOCATION_ID=5ad2b81f504d4d4493366f364d13f297" "JOURNAL_STREAM=9:358624446" "META_BUILDLET_BINARY_URL=https://storage.googleapis.com/go-builder-data/buildlet.linux-s390x" "GO_STAGE0_NET_DELAY=0s" "GO_STAGE0_DL_DELAY=100ms" "WORKDIR=/data/golang/workdir" "GOROOT_BOOTSTRAP=/data/golang/workdir/go1.4" "GO_BUILDER_NAME=linux-s390x-ibm-race" "GO_BUILDER_FLAKY_NET=1" "GO_TEST_TIMEOUT_SCALE=5" "GOBIN=" "TMPDIR=/data/golang/workdir/tmp" "GOCACHE=/data/golang/workdir/gocache" "GOPLSCACHE=/data/golang/workdir/goplscache" "PWD=/data/golang/workdir/go/src"] in dir /data/golang/workdir/go/src

What did you do?

https://build.golang.org/log/5e29be06bbb7a8a4923d83a58972aa2a10677a94

What did you expect to see?

All tests passing.

What did you see instead?

--- FAIL: TestCrashDumpsAllThreads (0.11s)
    crash_unix_test.go:145: found 2 instances of main.crashDumpsAllThreadsLoop; expected 4
…
FAIL
FAIL	runtime	46.511s

(attn @golang/s390x)

About this issue

Original URL
State: open
Created 7 months ago
Comments: 21 (3 by maintainers)

Most upvoted comments

@mknyszek @bcmills We had a look at this issue and we used git bisect to figure out the culprit commit.

Bad Commit Id: de5b418b which introducing failure of TestCrashDumpsAllThreads() We had a look at the file changes corresponding to the bad commit, we see that there are some changes to signal_unix.go file and we suspect that issue lies with the changes made to this file. We are further having look at the changes and we will update when we find more about the root cause of the failure.

srinivas-pokala on Dec 15, 2023

Since this seems to be very specific to the builder, in triage we’re wondering if anyone from @golang/s390x could start taking a look. It seems to be failing often enough that bisection should identify the culprit, and then that might just reveal the issue.

mknyszek on Dec 13, 2023