Ethan Hunter

I think I'll have to try to replicate this locally before I can understand more. Upon review of the code, it really does seem like oversized messages are being broadcast...

Hi, thanks for looking into this issue! Allowing tls over non-standard ports seems like a very reasonable feature. I think the best way to do this would be a new...

> @Spaceman1701 Yup, PagerDuty incidents is exactly the reason I raised this issue in the past. @SuperQ 👍 this is why my preference is for alertmanager's contract to be "For...

Just tagging this as related/similar to #4621

I'm wondering if this is still necessary given the recent improvements in Silencer performance. The big improvement relevant to this change is https://github.com/prometheus/alertmanager/commit/6ceb38d06d4fd6ecfbbd11510c7bdf591d92a917 which allows multiple queries to execute concurrently...

cc @grobinson-grafana or @gotjosh is there any interest in this PR? Or something I can do to make this easier to review?

I just updated the PR to include a new benchmark that's the same as `1_inhibition_rule,_X_inhibiting_alerts` that uses the `Equal` field so that the target labelset is only inhibit by the...

Ah, I typo-d the benchmark function name for `1_inhibition_rule,_1000_inhibiting_alerts_w/_random_match-4` and `1_inhibition_rule,_10000_inhibiting_alerts_w/_random_match-4` which explains the strange scaling there. I just pushed a fixed version and things look more logical: ``` $...

Just for completeness, here's a full 6 iteration run which compares the implementation on `main` with the two local changes we discussed in the CNCF slack: - `store.Alerts` replaced with...

One last commit addressing benchmark performance for now - I found that the different in the `_last_rule_matches` performance between this PR and the souped-up `main` branch version is mostly calls...