-
Notifications
You must be signed in to change notification settings - Fork 2k
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
.murdock: test 200 repeats of known flakey test #18396
Conversation
@MrKevinWeiss The test failed due to a syntax error. Murdock is idle, so it's good time to do some tests now. |
Heh, but I was not idle. I will try to hit this on the weekend when there is not much traffic. |
hmmm |
This must be the only PR which we want to fail and keeps passing... |
Mobi6 is down again so I will retry. |
Maybe ccache is too optimistic, so repeated runs yield same result but cache miss causes hash mismatch? |
Ya, that will be the next phase, disable ccache... |
I wonder if the failed results are getting overwritten with passing ones... |
I will wait until the speedup #18423 is in before continuing with this as the feedback cycle is super long. |
Theory-> something to do with gitcache (so far I have only seen it on esp bootloaders or packages or things that need to checkout other repos) stm pkgFrom #18433
esp bootloader
esp app ?(many mismatches binary + partitions)
|
Now trying to multiply by 200 submitted jobs using the get_app_board_toolchain_pairs
98ce4d7
to
f132ebf
Compare
Status update, we still see mismatches in the nightlies (cryptoauth package tests). This PR has been rebased and we will see if we can make it reproducible. |
Ugg, this is still an issue but probably does not make sense to work on it during hacknack. |
Hmmm I give up on this. It seems like all the changes are somehow working through it. |
This is only a test to try to solve the occasional unrelated hash failure problem forcing CI reruns.