Skip to content

Commit

Permalink
[CI] Adjust CPU and memory limits
Browse files Browse the repository at this point in the history
After switching to n2-highcpu-32 machines, the CPU and memory
requests/limits were left at the same values. This led to pods not
being scheduled or failing due to running out of memory relative to
the requests.

This patch updates the values to reflect the actual machine shapes
that we are running on.
  • Loading branch information
boomanaiden154 committed Dec 12, 2024
1 parent cd03d11 commit 489e9fe
Show file tree
Hide file tree
Showing 3 changed files with 6 additions and 6 deletions.
6 changes: 3 additions & 3 deletions premerge/linux_container_pod_template.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -20,7 +20,7 @@ spec:
# It's not used for scheduling, but can be used by k8 for OOM kill.
requests:
cpu: "100m"
memory: "50Gi"
memory: "25Gi"
limits:
cpu: 56
memory: "100Gi"
cpu: 32
memory: "32Gi"
4 changes: 2 additions & 2 deletions premerge/linux_runners_values.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -47,10 +47,10 @@ template:
# itself doesn't need much, just using something enough not to get
# OOM killed.
requests:
cpu: 50
cpu: 25
memory: "2Gi"
limits:
cpu: 56
cpu: 32
memory: "2Gi"
env:
- name: ACTIONS_RUNNER_CONTAINER_HOOKS
Expand Down
2 changes: 1 addition & 1 deletion premerge/windows_runner_values.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -21,7 +21,7 @@ template:
- name: runner
resources:
requests:
cpu: 50
cpu: 25
# TODO(boomanaiden154): Update this image when we have a proper windows
# image ready.
image: ghcr.io/keenuts-test-org/windows-ci-image:latest
Expand Down

0 comments on commit 489e9fe

Please sign in to comment.