I have a GCP cloudRun container which I have configured with the following values for the startup probe (I confirmed they are correctly set on the revision)
- initial delay 100s
- timeout 10s
- failure threshold 30
- interval 10s
Now the problem I have is that the container still does not startup correctly and in the console I find the error
STARTUP HTTP probe failed 7 times consecutively for container
This confuses me since I expected 30 attempts to be made, not just 7 – I have no idea where the 7 is coming from and why my configured failure threshold is ignored.
I triple checked the configuration in the GUI and I definetly have 30 as a failure threshold (for the liveness probe it’s 6 so that is also not where the 7 is coming from)
I’m grateful for any advice
after playing around a bit more the issue apparently was the 240s max startup duration. While the error message is misleading I was able to come to that conclusion because if I increase the initial delay to 200 I get the error
STARTUP HTTP probe failed 2 times consecutively for container "log-cleanups-1" on path "/api/checks/warmup". The instance was not started.
so it seems to be that the timeout was just reached