OpsGenie Integration uses wrong 'close' endpoint about alertmanager HOT 9 CLOSED

I have a tentative fix for this issue. Since I'm new to go, I'm having someone else give me some feedback before I kick over a PR.

from alertmanager.

Did you sent the PR, happy to review it (should be relatively small I think)

from alertmanager.

I didn't yet. Unit tests worked, but in testing it 'for real' and it doesn't seem to. I'll send it over when I can verify that it really really works. Here's the in process branch in my fork.

from alertmanager.

[ Quoting [email protected] in "Re: [alertmanager] OpsGenie Integra..." ]

I didn't yet. Unit tests worked, but testing it 'for real' and it doesn't seem to. I'll send it over when I can verify that it really really works. Here's the in process branch in my fork.

Looking at that branch and OpsGenie's docs, it should work. What happens if you
use an Id?

/Miek

Miek Gieben

from alertmanager.

Sorry, I didn't spot this issue and @jbarratt's branch before I send my PR over in #124. Anyway, looking at jbarratt's branch, I can't see anything wrong with it - it does pretty much the same thing as I do in #124, and I've seen #124 work fine when deployed here - alerts are created and closed in OpsGenie as they should be.

from alertmanager.

Ah yes, noticed that too. I'm fine with either, but I prefer how master...jbarratt:opsgenie_endpoint_fix tests for the expected endpoint directly in the HTTP handler. Maybe adopt that?

from alertmanager.

Sure, can do. =)

from alertmanager.

I verified that this works for me. I think this can be closed.

from alertmanager.

Great!

from alertmanager.