action #173869
closed[BCI] bash segfaults in BCI testing on powerPC hosts
100%
Description
Observation¶
openQA test in scenario sle-16.0-BCI-Updates-ppc64le-bci-base_16.0_on_SLES_12-SP5@ppc64le fails in
root_BCI-tests_all
Given the fact that we are running SLFO code stream on old PowerPC workers, the BCI test containers are failing with segfaults when running binaries in the containers.
Dec 06 04:16:06.748541 linux-kmed systemd-coredump[8079]: Process 8055 (bash) of user 0 dumped core.
Stack trace of thread 1:
#0 0x00007fff8cedf934 n/a (ld64.so.2)
Dec 06 04:16:06.860660 linux-kmed dockerd[3204]: time="2024-12-06T04:16:06.860314464-05:00" level=error msg="Error setting up exec command in container b15da49412efd44688bafd6b6adc495bbd7bdfb0e7337c0b66aa5b1eba82c26b: container b15da49412efd44688bafd6b6adc495bbd7bdfb0e7337c0b66aa5b1eba82c26b is not running"
As in SLM testing, we need to use emulated PowerPC workers on SLEM16, anything newer than POWER8.
Expected result¶
Last good: None
Acceptance criteria¶
1) SLFO tests are using Power9 workers, currently only available as emulated workers
QEMU=ppc64
QEMUCPU=POWER9
WORKER_CLASS=qemu_x86_64-large-mem
MACHINE=ppc64le-emu
2) Make sure that host images are adjusted for the emulated workers
Updated by ph03nix 3 months ago
Trying to work out the status of Power9+ workers in openQA: https://suse.slack.com/archives/C02CANHLANP/p1733731071602019
Updated by ph03nix 3 months ago
To try out the new scheduling, I deleted all ppc64le jobs from https://openqa.suse.de/tests/overview?groupid=608&version=16.0&distri=sle&build=7.2_base-image and will now wait for the new jobs to be scheduled.
Updated by ph03nix 3 months ago
Waiting for new test runs in https://openqa.suse.de/tests/overview?build=7.2_base-image&groupid=608&version=16.0&distri=sle
Updated by ph03nix 3 months ago
- Related to action #174073: [BCI] Investigate kiwi failures on PPC added