You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently metrics exposed by Tetragon don't have a label with the node name. node label is added by a relabeling in the default ServiceMonitor. This works well enough, however:
In deployments not using the default ServiceMonitor, users have to manually configure scrape config to add such relabeling.
In some deployments (non-k8s in particular) node name might be not available to Prometheus at the scrape time.
node label on metrics might end up being different than node_name field in JSON events.
It's always nice to have metrics labels matching events fields.
So I'm suggesting to add node_name label to all metrics exposed by Tetragon - it should be retrieved in the same way as node_name in events.
The text was updated successfully, but these errors were encountered:
Currently metrics exposed by Tetragon don't have a label with the node name.
node
label is added by a relabeling in the default ServiceMonitor. This works well enough, however:node
label on metrics might end up being different thannode_name
field in JSON events.So I'm suggesting to add
node_name
label to all metrics exposed by Tetragon - it should be retrieved in the same way asnode_name
in events.The text was updated successfully, but these errors were encountered: