No other errors are in available, and the build does not even start a log. Sifting through logs explorer with no filter also doesn’t yield a smoking gun.
Retrying the build normally fixes it, although it can happen 2-3 times subsequently.
Any ideas of how to investigate this issue? Or common causes of this phenomenon?
Couple interesting things about the build:
It’s a GitHub repo which hosts the code and cloudbuild.yaml
It is run in a private pool. The private pool’s network is peered with a private GKE cluster network via VPN tunnel, as the build does some K8 deployment.
If the steps didn’t work, you may file a bug so that our engineers could take a look at this. We don’t have a specific ETA but you can keep track of its progress once the ticket has been created.
The repository configuration is fine, the issue is sporadic (30% of the time it fails) which isn’t really congruent with incorrect github configuration. I’ve have an open support case with google #50154986 which we’re investigating further.
This is not the same problem. My issue happens sporadically 30% of the time and is not likely related to a GitHub connection issue. I will post the solution here when I find it, but I doubt it will help you.
To follow up on this, it turned out it was due to quotas being hit on artifact registry requests. We have a maven build that does a lot of requests to the registry. It was easy to get these quotas bumped and we haven’t seen it since
