action #35044
closed[sle][functional][u][hyperv][easy] textmode@svirt-hyperv-uefi exceeds max job time - 3h
0%
Description
[2018-04-16T13:24:34.0918 CEST] [info] got job 1620812: 01620812-sle-12-SP4-Server-DVD-x86_64-Build0236-textmode@svirt-hyperv-uefi
[2018-04-16T13:24:34.0944 CEST] [info] OpenQA::Worker::Cache: Initialized with openqa.suse.de at /var/lib/openqa/cache, current size is 50604837888
[2018-04-16T13:24:35.0289 CEST] [info] Downloading SLE-12-SP4-Server-DVD-x86_64-Build0236-Media1.iso from openqa.suse.de/tests/1620812/asset/iso/SLE-12-SP4-Server-DVD-x86_64-Build0236-Media1.iso
[2018-04-16T13:25:46.0693 CEST] [info] CACHE: updating the /var/lib/openqa/cache/SLE-12-SP4-Server-DVD-x86_64-Build0236-Media1.iso with "e6600000-56907704bfd60" and 3865051136
[2018-04-16T13:25:50.0024 CEST] [info] 19791: WORKING 1620812
[2018-04-16T16:25:50.0014 CEST] [warn] max job time exceeded, aborting 01620812-sle-12-SP4-Server-DVD-x86_64-Build0236-textmode@svirt-hyperv-uefi
[2018-04-16T16:25:51.0429 CEST] [info] uploading consoletest_setup-loadavg_consoletest_setup.txt
"MAX_JOB_TIME" : "10800" is not enough for textmode@svirt-hyperv-uefi job or the job is just too big for hyperv machine.
Observation¶
openQA test in scenario sle-12-SP4-Server-DVD-x86_64-textmode@svirt-hyperv-uefi fails in
curl_https
Reproducible¶
Fails since (at least) Build 0152
Acceptance criteria¶
- AC1: A higher timeout is applied for hyperv jobs by default
Suggestions¶
I guess it makes sense to scale the MAX_JOB_TIME based on TIMEOUT_SCALE. Challenge is that TIMEOUT_SCALE belongs to os-autoinst, MAX_JOB_TIME belongs to openQA but it should be ok to look for the variable existing in the context of os-autoinst and scale it accordingly.
What we can do still is apply a higher default timeout in the machine setting for hyperv - unless that overwrites settings in the test suite but that could be simply tried by checking that with a local openQA instance.
Further details¶
Always latest result in this scenario: latest