action #174919
closedsome qemu OSD tests on worker30.oqa.prg2.suse.org fail with "Parameter 'capability' does not accept value 'compress'" due to undocumented use of "virtualization" repo
0%
Description
Observation¶
From https://suse.slack.com/archives/C02CANHLANP/p1735552360309679
reproducible failure in x86_64 . Job finishes always with result incomplete ( example https://openqa.suse.de/tests/16314521 )
error is backend died: QMP command migrate-set-capabilities failed: GenericError; Parameter 'capability' does not accept value 'compress' at /usr/lib/os-autoinst/backend/qemu.pm line 1105. .
Is this known ? (edited)
seems the problem was actually that for undocumented reasons the "virtualization" repo was deployed on worker30
Updated by okurz about 2 months ago
- Copied from action #167218: Worker with qemu 9.1.0 (Tumbleweed) fails in Incomplete state - Parameter 'capability' does not accept value 'compress' added
Updated by okurz about 2 months ago
- Related to action #157975: Upgrade osd workers to openSUSE Leap 15.6 size:S added
Updated by okurz about 2 months ago
- Due date set to 2025-01-17
- Status changed from In Progress to Feedback
- Priority changed from High to Normal
I think the problem was triggered by #157975
I did zypper rr Virtualization && zypper dup && reboot and called
WORKER=worker30 comment="poo#174919" failed_since=2024-12-25 host=openqa.suse.de openqa-advanced-retrigger-jobs
Updated by gpathak about 2 months ago
The jobs are still failing: https://openqa.suse.de/tests/16353444#step/installation/12
Updated by livdywan about 2 months ago
So just to track the relevant context here:
- https://github.com/os-autoinst/os-autoinst/pull/2608 was merged, and has the capability / multifd support - this was part of #167218
- worker30 still uses the older qemu version
seems the problem was actually that for undocumented reasons the "virtualization" repo was deployed on worker30
this repo has the new qemu. gaurav added it because it doesn't seem to be available anywhere else. me thinks we should check if it's in factory, and otherwise see to it being added.
Updated by okurz about 2 months ago
OSD should run stable Leap so if any updates are needed for anything besides the scope of this ticket then such updates should be submitted to Leap
Updated by okurz about 2 months ago
- Due date deleted (
2025-01-17) - Status changed from Feedback to Resolved
from https://openqa.suse.de/tests/latest?arch=x86_64&distri=sle&flavor=Server-DVD-Updates&machine=64bit&test=publiccloud_tools_cli_azure&version=15-SP6 we can see stable and passed jobs again. w30 is on the same state as other osd workers and consistent.
Updated by jbaier_cz about 1 month ago
- Related to action #170296: [openqa-in-openqa][sporadic] test fails in test_running - ping_client is not complete size:S added