Nikolay Martyanov

Leave it for now, please. I can use it as one of the references. Maybe we can add some tag to the PR.

Oh... I have a looot to add to the upgrade tests... But first I have to understand on how to run them))

This code has just completely removed my ~/projects, with all the projects inside (tens of them). I demand this code to be eliminated!!! (Well, if the code is fixed and...

https://github.com/lf-edge/eve/actions/runs/10820461464?pr=4239 test_suite_pr (amd64, kvm, generic) / Storage test suite (ext4): tests/volume/testdata/volumes_test.txt#L60 [stdout] Reboot Test === RUN TestReboot reboot_test.go:162: time: 2024-09-12T06:21:16.315099837Z out: Wait for state of b23cb68b-686a-4bb1-a9c4-dafe4d101b3b testContext.go:380: done waiting for...

> Also, we should be careful about what we label as flaky failure. Just because a test fails intermittently it does not necessary mean that the infrastructure is to blame....

> @OhmSpectator The `reboot or kernel panic due to watchdog or kernel bug (no kdump)` is affecting any test. The EVE VM is likely not getting CPU time from the...

Argh... Are the artifacts lost, when running a new attempt? =( https://github.com/lf-edge/eve/actions/runs/10820461464/attempts/2?pr=4239

> @OhmSpectator The `reboot or kernel panic due to watchdog or kernel bug (no kdump)` is affecting any test. The EVE VM is likely not getting CPU time from the...

> Yes, the `WATCHDOG` event suggests issue with the runner. If something in EVE actually got stuck, our software watchdog would trigger sooner. Good to know, thanks!

Interesting. If there is a very clear criterion (https://github.com/lf-edge/eve/issues/4246#issuecomment-2345759309) for filtering out test failures caused by the watchdog error, could we mark these failures correspondingly in the test result? @yash-zededa,...