feat: make alerts configurable #19
Reference in New Issue
Block a user
No description provided.
Delete Branch "configurable-alerts"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Alert threshholds can now be configured via env variables
I also seperated the basic auth, so that it is possible to deploy only prometheus without promtail
@dannygroenewegen did you get around to test this yet? :)
nice!
@ -168,1 +161,4 @@It is possible to enable the following alerts, by uncommenting the corresponding env variable:- node disk space: `ALERT_NODE_DISK_SPACE_LEFT`- node memory usage: `ALERT_NODE_MEMORY_USAGE`nit: should probably remove the options from here since they are explained in the env file, otherwise we would have to remember that these need to be updates in 2 places
Ah yes, good point!
@ -80,4 +81,2 @@#GF_MATRIX_HOMESERVER_URL="<homeserver-url>"# ALerts#ALERT_BACKUP_FAILED_ENABLED=trueI've seen mentions that these backup alerts don't work, but is that the case for everyone? Or could someone still be relying on this?
And if these envs are removed, then grafana-alerts.json.tmpl can also be removed?
@ -87,1 +83,3 @@#ALERT_NODE_MEMORY_USAGE_ENABLED=true## ALerts# Node disk space alert will trigger when free disk space left is below the given number in percentGood enough for now. But 10% of a 20GB disk or a 2TB disk is a bit different. Can imagine we want a bit more fine-grained control depending on disk size and usage.
View command line instructions
Checkout
From your project repository, check out a new branch and test the changes.