-
Notifications
You must be signed in to change notification settings - Fork 21
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Tast run failures because of Tast internals #838
Comments
I'll try to find time and work on fixing this failure. |
@a-wai @padovan @nuclearcat @laura-nao I've confirmed that not even a single tast test (from kernel, hardaware, platform, power and mm-misc) is able to run because we get the error from Tast tool, "Form Factor not known". It is a known problem which we used to hit only on some jobs. We were able to get results on other runs. But from several days, there are no results. I was thinking maybe we are not getting successful run results in KCIDB as lava is unable to submit them and hence Grafana/Result-Summary are unable to show me results. Now that Lava issue has been resolved, I've again found no result today in Grafana/Result-summary. So I've looked at logs of each I'm able to find video/codec tast tests getting executed without an issue. Not sure why. |
FWIW, while testing ARM64 devices on R130 I noticed 6.12-rc and -next kernels seemed to always fail, but 6.11 (or staging-stable, which is tracking 6.1) seemed to be passing. Maybe a change in a Kconfig option upstream which breaks our setup in a different and creative way? Disclaimer: my sample size was quite small there, so it might just be coincidence, further analysis is needed. |
For stable-rc, 11-20 October we were getting usual good results. From 21-31, we have only got some results on Qualcomm targets. All other targets have "Form factor error". I've not looked specifically at kernel versions. This could be the problem with kernels or configs. But stable kernels seem affected as well. They shouldn't have gotten affected if there was new change to the recent mainline kernels. But we can never be sure until we find root cause and fix it. |
I think we should report these errors and ask the original tast developers about this situation. Over the next few days, I can collect more results from R130 chromiumos. Maybe we'll better or even worse situation. |
cc: @laura-nao |
I've seen so many of these kind of errors. Strangely tast outputs errors and no test is executed. This must be looked at and fixed. This is the number 1 reason most tast jobs fail and get marked as infrastructure failures.
https://lava.collabora.dev/scheduler/job/16060973
https://lava.collabora.dev/scheduler/job/16281873#L9431
It is a known long-standing issue.
b#334788335
cc: @laura-nao @a-wai @padovan @nuclearcat
The text was updated successfully, but these errors were encountered: