QA (public) » openQA Project (public) » openQA Tests (public)

happening in SLE15 GMC.

SR to patch upstream bug: https://build.opensuse.org/request/show/616259
SR to patch openSUSE bug: https://build.opensuse.org/request/show/616468

Investigating s390x issue.

At the moment SLE15 is already on GMC phase. Adjusting prio.

I cannot investigate further due to a malfunction of the s390x-susekvm worker: https://progress.opensuse.org/issues/33202

I forgot to change status after s390x worker was self-healed

The problem is identified.

SLE15 is GM, so lowering priority.

So I discussed in detail with slindomansilla related to #34996 as well. AFAIU this ticket is for SLE in particular. Currently we focus on SLE12SP4 and for SLE15 the tests had been failing for a long time. This is why I delay the ticket plan to M20 as we should get openSUSE Tumbleweed in a stable state first and then revisit the situation for SLE tests. Whenever we see tests failing consistently and for a long time we should move the test scenario to the according "test development" job groups on the openQA server(s). Let's focus into #34996 . @tsaupe of course the plan is only according to our planning as QSF (QA SLE functional) to look into this issue again later, milestone 20, which is 2018-11, but you are free to continue on this ticket as you like :)

Seems like no real progress here. By now I think we should take a look into SLE15 tests, at least remove the red blob from the validation tests. https://bugzilla.suse.com/show_bug.cgi?id=1105101 is currently used as label in these tests but also there nothing is moving forward, multiple unattended openqa-review reminder comments.

A SR provided by tsaupe was accepted yesterday: https://build.suse.de/request/show/175568

I will verify the fix on OSD.
Waiting for:

• aarch64: https://openqa.suse.de/tests/2240805
• s390x: https://openqa.suse.de/tests/2240806

The job times out. I will try to reproduce how much time it needs to adapt the test.

PR to increase timeout: https://github.com/os-autoinst/os-autoinst-distri-opensuse/pull/6200

Both fail. 1 hour timeout is not enough. Investigating...

Waiting for verification on OSD:

• aarch64: https://openqa.suse.de/tests/2258586
• s390x: https://openqa.suse.de/tests/2258587

Results for build 96.7:

• aarch64: https://openqa.suse.de/tests/2261889#step/systemd_testsuite/16 (fails, it still hangs on TEST-02)
• ppc64le: https://openqa.suse.de/tests/2261978#step/systemd_testsuite/53 (it fails now in TEST-10)
• s390x: https://openqa.suse.de/tests/2263572
• x86_64: https://openqa.suse.de/tests/2262698 (passes)

Investigating fail on TEST-10:

• Try to reproduce in the previous build (95.1): https://openqa.suse.de/tests/2263603
• Try to reproduce in the current build (96.7): https://openqa.suse.de/tests/2263604

New SR from tblume to fix aarch64 problem: https://build.suse.de/request/show/177815

Waiting for verification: https://openqa.suse.de/tests/2263888

M20 is over, as we already shifted a lot I am not sure how this will go on. As we still have jobs failing labeled with this ticket, e.g. https://openqa.suse.de/tests/2271738#step/systemd_testsuite/17, I see it as urgent now. The urgency could be handled by removing the scenario again. But I leave it to you to decide which way to go, just … urgently please :)

At the moment there is no s390x worker to verify.
OSD ones are getting freeze.
QSF shared workers are not working because the s390x zkvm backend was never adapted to work on workers with enabled cache.

Fixing s390x backend first and moving aarch64 and ppc64le scenarios to development job group: https://openqa.suse.de/admin/job_templates/96

• A fix was applied for XEN: https://github.com/os-autoinst/os-autoinst-distri-opensuse/pull/6191
• That needs to be applied for s390x: https://github.com/os-autoinst/os-autoinst-distri-opensuse/pull/6253

Urgency removed after moving aarch64 and ppc64le scenarios to development job group.

@okurz, feel free to set normal prio.

yes, normal prio is fine.

Blocked by: https://gitlab.suse.de/qa-sle/shared-development-worker/issues/1

Worker is working again: http://slindomansilla-vm.qa.suse.de/tests/959

s390x verification failed: https://openqa.suse.de/tests/2288353

Waiting for tsaupe's feedback on https://build.suse.de/request/show/177956

Not able to verify on s390x until there is an available qcow2 image of SLE15-SP1 textmode for s390x-kvm-sle12

Blocked by #45158

due to blocker

Issue not reproducible anymore