action #151399
Updated by okurz about 1 year ago
## Motivation With #109920 openqa-investigate can identify clear *PRODUCT REGRESSIONS* and with #132272 we identify *TEST REGRESSIONS*. We can extend that same concept to now clearly identify infrastructure regressions and inform us about it ## Acceptance criteria * **AC1:** On failed openQA jobs with openqa-investigate info "S0: retry X, last_good_test X, last_good_build X, last_good_test+build X" (X: failed, V: passed) a comment is written pointing to a likely *infrastructure* regression * **AC2:** No such comment is written on other jobs * **AC3:** The tools team receives a notification about such issues ## Suggestions * Learn what had been done in #109920 for product regressions and #132272 for test regressions * Do what had been done in https://github.com/os-autoinst/scripts/pull/223 and others but accordingly for infrastructure issues * Decide on a way how to send notifications for such issues, e.g. something that could be read and alerted on within grafana or sending an email in a special mailing list and/or Slack channel although I guess for those we would receive too many messages to be useful. How about just telegraf doing an SQL query for comments with the specific message? useful ## Steps * https://github.com/os-autoinst/scripts/pull/265 Identify reproducible test issues