Azure pipelines looks unhealthy for epoch branches #21200
Comments
|
@stephenmcgruer Looked at Windows VMs and they all appear to be in a working state. However, i do see a bunch of errors to connect to github: https://dev.azure.com/web-platform-tests/wpt/_build/results?buildId=40180&view=logs&j=425415bf-d1d5-5063-08bf-7eadf688f647&t=07c0114d-1a67-59a4-150d-697f6f86a11d |
|
@mustjab I see that too, but I doubt it's a GitHub issue. Other branches on Azure aren't having the problem (epochs/daily looks fine, sampled a few random PR runs and they seem fine, ...) and I'm not aware of any other reports of GitHub issues. Could there be a network problem with the machines assigned to |
|
@stephenmcgruer You're correct, one of the VMs in Windows pool (w10c00000J) is having network issues and was the cause for most of the failures. I deleted that VM, so we should see future runs pass. If we find that there are other machines in the pool that have these issues, then i'll open a ticket with Azure and will try to re-create the VM pool again to see if we can workaround this issue. |
|
Looks like latest three_hourly run has passed. |
|
Yes, I think we're good. There were some 500s from wpt.fyi over the last few days, but that's probably on our end to start looking at. Thanks for your work here! |
I noticed because the stable runs check went red; Safari is missing for bd72596; https://wpt.fyi/runs?label=master&label=stable&max-count=100&product=chrome&product=firefox&product=safari
Looking at https://dev.azure.com/web-platform-tests/wpt/_build?definitionId=1&_a=summary&view=branches, that run failed but also runs on the three_hourly have been failing for a while.
The epochs/daily run failed with:
##[error]We stopped hearing from agent Azure Pipelines 5. Verify the agent machine is running and has a healthy network connection. Anything that terminates an agent process, starves it for CPU, or blocks its network access can cause this error. For more information, see: https://go.microsoft.com/fwlink/?linkid=846610https://dev.azure.com/web-platform-tests/wpt/_build/results?buildId=40111&view=logs&j=e8032fe0-b3da-5a76-081d-ec1db1bea50b
The epochs/three_hourly runs also seem to fail with the same error, and some of them have a similar error for Edge jobs (
##[error]We stopped hearing from agent w10c00000D. Verify the agent machine is running and has a healthy network connection. Anything that terminates an agent process, starves it for CPU, or blocks its network access can cause this error. For more information, see: https://go.microsoft.com/fwlink/?linkid=846610)@mustjab can you have a look thanks?
The text was updated successfully, but these errors were encountered: