action #59226
closed[kernel]ltp][spvm][sporadic] test fails in boot_ltp - cat /tmp/ver_linux_before.txt timeout
0%
Description
Observation¶
openQA test in scenario sle-15-SP2-Installer-DVD-ppc64le-ltp_cve_spvm@ppc64le-spvm fails in
boot_ltp
Test suite description¶
Reproducible¶
Fails since (at least) Build 77.1
Expected result¶
Last good: 77.1 (or more recent)
Further details¶
Always latest result in this scenario: latest
Randomly happens (but often), timeout increase will be probably needed.
Updated by jlausuch about 5 years ago
I've seen many timeout failures with script_output. I think we just need to add a timeout. So, something like
my $ver_linux_out = script_output("cat $ver_linux_log", 300);
Updated by pcervinka about 5 years ago
- Status changed from New to In Progress
- Target version changed from 454 to 445
Updated by pcervinka about 5 years ago
Had issues with my openQA instance with spvm in last 2+days and couldn't progress on the case, but my findings are now:
- happens for ipmi as well https://openqa.suse.de/tests/3554259#step/boot_ltp/23 (notified pvorel as well)
- simple solution with timeout increase is not correct one
- when it works it usually takes 2 seconds and when looking on logs, all was printed, file is quite small, there is no reason to take it more
Updated by pcervinka about 5 years ago
Failing even with timeout change to 180:
[2019-11-13T13:35:56.395 CET] [debug] /var/lib/openqa/share/tests/sle/tests/kernel/boot_ltp.pm:78 called testapi::script_output
[2019-11-13T13:35:56.395 CET] [debug] <<< testapi::type_string(string='(echo ORQ2Z; bash -eox pipefail /tmp/scriptORQ2Z.sh ; echo SCRIPT_FINISHEDORQ2Z-$?-) | tee /dev/sshserial
', max_interval=250, wait_screen_changes=0, wait_still_screen=0, timeout=30, similarity_level=47)
[2019-11-13T13:35:59.007 CET] [debug] /var/lib/openqa/share/tests/sle/tests/kernel/boot_ltp.pm:78 called testapi::script_output
[2019-11-13T13:35:59.007 CET] [debug] <<< testapi::wait_serial(quiet=undef, no_regex=0, regexp='SCRIPT_FINISHEDORQ2Z-\\d+-', timeout=180, record_output=1, expect_not_found=0, buffer_size=undef)
ORQ2Z
Updated by pcervinka about 5 years ago
- Status changed from In Progress to Feedback
I was successfully reproducing failure http://10.100.12.105/tests/3512 with os-autoinst version 4.6.1572372410.447dab86-lp151.214.1. When updated to 4.6.1573588068.2f8e74c9-lp151.234.1 it started work without failures. Did verification with multiple cat commands http://10.100.12.105/tests/3517. Jobs on osd seems to be fine now https://openqa.suse.de/tests/3588069.
Seems to be related to fix https://github.com/os-autoinst/os-autoinst/pull/1258.
Let's see more test runs.
Updated by pcervinka about 5 years ago
- Related to action #59190: Broken SSH serial console added
Updated by jlausuch about 5 years ago
pcervinka wrote:
I was successfully reproducing failure http://10.100.12.105/tests/3512 with os-autoinst version 4.6.1572372410.447dab86-lp151.214.1. When updated to 4.6.1573588068.2f8e74c9-lp151.234.1 it started work without failures. Did verification with multiple cat commands http://10.100.12.105/tests/3517. Jobs on osd seems to be fine now https://openqa.suse.de/tests/3588069.
Seems to be related to fix https://github.com/os-autoinst/os-autoinst/pull/1258.
Let's see more test runs.
Nice! I hope it also applies for ipmi jobs too :) I remember similar errors when I was using script_output in ipmi jobs.
Thanks for doing these checks.
Updated by pcervinka about 5 years ago
- Status changed from Feedback to Resolved
Works fine, there are no more issues on spvm.
Updated by jlausuch almost 5 years ago
- Target version changed from 445 to 457
yes, latest builds seem ok
https://openqa.suse.de/tests/3778922#next_previous
Updated by pcervinka about 4 years ago
- Target version changed from 457 to QE Kernel Done