action #163769
closedcoordination #102915: [saga][epic] Automated classification of failures
coordination #166655: [epic] openqa-label-known-issues
Errors in gru journal: jq (173 /opt/os-autoinst-scripts/openqa-label-known-issues): parse error: Invalid numeric literal at line 1, column 8 (rc: 4 Input: >>>Waiting 1s until retry #12
0%
Description
Observation¶
On osd and o3 I see many of those:
(Output cut off on the right side due to the very long JSON)
Jul 11 13:31:13 openqa openqa-gru[9530]: /opt/os-autoinst-scripts/_common: line 62: echo: write error: Broken pipe
Jul 11 13:31:13 openqa openqa-gru[9529]: /opt/os-autoinst-scripts/_common: ERROR: line 62
Jul 11 13:31:13 openqa openqa-gru[8339]: /opt/os-autoinst-scripts/_common: ERROR: line 62
Jul 11 13:31:13 openqa openqa-gru[8339]: jq (173 /opt/os-autoinst-scripts/openqa-label-known-issues): parse error: Invalid numeric literal at line 1, column 8 (rc: 4 Input: >>>Waiting 1s until retry #12
Jul 11 13:31:13 openqa openqa-gru[8339]: {"issues":[{"id":162596,"project":{"id":125,"name":"openQA Infrastructure"},"tracker":{"id":4,"name":"action"},"status":{"id":15,"name":"Blocked","is_closed":false},"pr>
Jul 11 13:31:13 openqa openqa-gru[8339]: ook 4.02 seconds for 35 candidate needles - make your needles more specific\r\n [2021-08-03T07:25:03.632 CEST] [debug] no match: -0.5s, best candidate: linux-login-2>
Jul 11 13:31:13 openqa openqa-gru[8339]: ts-Kernel-KOTD@64bit-with-ltp.qcow2\" successful, new cache size is 49GiB\r\n\r\n[2020-10-13T10:52:32.0261 CEST] [debug] [pid:36934] Found ISO, caching SLE-15-SP1-Insta
It looks like the Waiting 1s until retry #12
is added to the output that jq considers as its input.
Suggestions¶
- Print the "Waiting ..." message to stderr instead of stdout. We already have a
warn
function for that.
Updated by tinita 5 months ago
- Status changed from In Progress to Feedback
https://github.com/os-autoinst/scripts/pull/327 Print warning messages to stderr
Updated by tinita 5 months ago
Looked into the log.
The warning now goes to stderr, so that issue is gone.
Now I'm seeing some other weird issue (that was already there before):
Jul 11 15:15:07 openqa openqa-gru[7683]: curl (173 /opt/os-autoinst-scripts/openqa-label-known-issues): Error fetching (-L --user-agent openqa-label-known-issues -sS https://progress.opensuse.org/projects/openqav3/issues.json?limit=200&subproject_id=*&subject=~auto_review%3A): {"issues":[{"id": ....... very long JSON output
So apparently the request was successful but it says it wasn't.
Maybe this should be a new ticket.
Updated by tinita 5 months ago
Oh, I see.
The JSON output ends with:
...{"id":135839,"project":{"id":36,"name":"openQA Tests"},"tracker":{"id":4,"name":"action"},"status":{"id":1,"name":"New","is_closed":false},"priority":{"id":4,"name":"Normal"},"author":{"id":17668,"name":"okurz"},"category":{"id":152,"name":"Bugs in existing tests"},"subject":"[desktop] test fails in login_test auto_review:\"no candidate needle.*gnome-settings\":retry","description":"## Observation\r\n\r\nopenQA test in scenario opensuse-Tumbleweed-DVD-x86_64-desktcurl: (18) Recv failure: Connection reset by peer
So redmine returned a partial response.
The output is annoyingly long and it's hard to find the actual failure. But if we cut off the end, we won't see it, but we also want to see the start. Maybe the content should be cut in the middle.
Updated by livdywan 4 months ago
- Copied to action #164153: Errors in gru journal: openqa-label-known-issues error fetching JSON output size:S added