Potential issue when sending SIGHUP before ready

[juliusv]

From a user on IRC:

The user reported Prometheus blocking on startup. Then:

"I was able to find the root cause of the problem. It was related to my program sending a SIGHUP to prometheus to reload it's configuration before prometheus registered the SIGHUP handler"

We should investigate whether our startup HUP signal handling is broken.

[gouthamve]

If someone wants to look into this, here are some pointers:
Out HUP handler is registered just before this: https://github.com/prometheus/prometheus/blob/master/cmd/prometheus/main.go#L188

Since it is hard to actually send a HUP before stuff starts, I think the best thing we can do is to add time.Sleep() at that point and see if something is off.
Then move the sleep to after localstorage starts. After notifier.Start() and so on to see if anything breaks anywhere.

If nothing breaks, I think it is safe to close this.

[alwindoss]

I am willing to work on this issue.
I am new to prometheus and relatively new to "go" and have worked on some hobby projects. Could someone help me through the initials here.

[grobie]

@krasi-georgiev can you confirm this is fixed with the main.go refactoring? This would be a great test.

[krasi-georgiev]

yes just tested with a sleep before close(reloadReady) and it doesn't block the startup

will add it to the main.go tests issue

[lock]

This thread has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.