action #163475
opencoordination #102906: [saga][epic] Increased stability of tests with less "known failures", known incompletes handled automatically within openQA
coordination #102912: [epic] Simplify investigation of job failures - 2nd
On-demand diff'ing of any result+log file from failed openQA jobs
0%
Description
Motivation¶
The #investigation tab on openQA job details already provides a convenient diff'ing of vars.json files and for the optionally provided "diff_packages_to_last_good" and such. Test reviewers like the investigation tab and would like have access to convenient diff'ing for other files as well, e.g. all result+log files that are present for both "last good" and "first bad". We should offer on-demand diff'ing of any result+log files that are present in both types of job, e.g. collapsed entries that could be clicked to trigger the actual evaluation on click.
Acceptance criteria¶
- AC1: A difference for all result+log files that are present for both "last good" and "first bad" is available in the investigation tab on request
- AC2: The investigation tab is still readable enough and not too long
- AC3: The investigation tab showing only the main information, e.g. diff of settings, still renders quickly enough
Suggestions¶
- Read the current implementation in https://github.com/os-autoinst/openQA/blob/136b92a59433850f8bf077594139d420fcf2c3d4/lib/OpenQA/Schema/Result/Jobs.pm#L1885
- Extend tests in https://github.com/os-autoinst/openQA/blob/136b92a59433850f8bf077594139d420fcf2c3d4/t/10-jobs.t#L436
- Dynamically compile list of all result&log files that are common in both jobs used for comparison
- Consider using dynamic triggering of actions based on clicks on investigation table entries
- Monitor for performance impact
No data to display