Add "safe-to-evict" autoscaler annotations for components with emptyDir
mounted
#676
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description
Changes proposed in this pull request:
emptyDir
mountedPrometheus, Alertmanager and Neo4j have
emptyDir
mounted which prevent them to scale down clusters that use Cluster Autoscaler, e.g. our long-running one on GCP:While I didn't see such errors for kubed, argo server and grafana, they also use
emptyDir
for cache, so I annotated them just to be safe.Notes
If we upgraded to GKE 1.22 we wouldn't need such annotations on the pods:
https://cloud.google.com/kubernetes-engine/docs/release-notes#October_27_2021
There's already open task for that: #611
Testing
Run
make dev-cluster
and observe new annotations:Open the file in a text editor and verify that for all pods with
emptyDir
the annotationcluster-autoscaler.kubernetes.io/safe-to-evict: "true"
is properly set.Related links