Increase time waiting for kubelet #205
Conversation
WalkthroughIncreased a wait_for timeout for kubelet port 6443 from 300s to 600s and updated the base image digest in Changes
Sequence Diagram(s)(No sequence diagram generated — changes are parameter and image-digest updates without control-flow or new feature interaction.) Estimated code review effort🎯 2 (Simple) | ⏱️ ~10 minutes Possibly related PRs
Pre-merge checks (3 passed)✅ Passed checks (3 passed)
Poem
Tip 👮 Agentic pre-merge checks are now available in preview!Pro plan users can now enable pre-merge checks in their settings to enforce checklists before merging PRs.
Please see the documentation for more information. Example: reviews:
pre_merge_checks:
custom_checks:
- name: "Undocumented Breaking Changes"
mode: "warning"
instructions: |
Pass/fail criteria: All breaking changes to public APIs, CLI flags, environment variables, configuration keys, database schemas, or HTTP/GraphQL endpoints must be documented in the "Breaking Change" section of the PR description and in CHANGELOG.md. Exclude purely internal or private changes (e.g., code not exported from package entry points or explicitly marked as internal).Please share your feedback with us on this Discord post. ✨ Finishing touches🧪 Generate unit tests
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 0
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
ansible/roles/deploy-crc-cloud/tasks/kubelet.yaml (1)
20-23: Fix premature success inuntilcondition (use AND, not OR).With ORs, the first run can pass when stdout is empty, then fail due to
failed_when. Use ANDs so all error markers are absent and resources appear before proceeding.until: | - ('No resources found' not in _openshift_containers.stdout) or - ('connect: connection refused' not in _openshift_containers.stderr) or - ('get current server API group list' not in _openshift_containers.stderr) + ('No resources found' not in _openshift_containers.stdout) and + ('connect: connection refused' not in _openshift_containers.stderr) and + ('get current server API group list' not in _openshift_containers.stderr)
🧹 Nitpick comments (1)
ansible/roles/deploy-crc-cloud/tasks/kubelet.yaml (1)
10-14: Make wait parameters configurable.Expose timeout/delay as vars so environments can tune without code changes.
- name: Wait for port 6443 to be up ansible.builtin.wait_for: - timeout: 600 + timeout: "{{ kube_apiserver_wait_timeout | default(600) }}" port: 6443 - delay: 60 + delay: "{{ kube_apiserver_wait_delay | default(60) }}"
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (1)
ansible/roles/deploy-crc-cloud/tasks/kubelet.yaml(1 hunks)
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (2)
- GitHub Check: build (macOS-latest, 1.20)
- GitHub Check: build (ubuntu-latest, 1.20)
🔇 Additional comments (1)
ansible/roles/deploy-crc-cloud/tasks/kubelet.yaml (1)
12-12: LGTM: Timeout increased to 600s.The change directly addresses slow startups and is safe/idempotent.
Sometimes, 300 seconds waiting time for port :6443 to be available is not enough. Increase the time to 600 seconds. Also update CentOS 9 Stream container SHA-256 to unlock the CI. Signed-off-by: Daniel Pawlik <[email protected]>
danpawlik
force-pushed
the
increase-waiting-for-kubelet
branch
from
c1d9174 to
d715f31
Compare
There was a problem hiding this comment.
Actionable comments posted: 0
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
ansible/roles/deploy-crc-cloud/tasks/kubelet.yaml (1)
21-23: Likely logic bug: use AND instead of OR inuntil.With OR, the loop can exit early if any single error string isn’t present, even when others still are. AND ensures all error indicators are absent before proceeding.
- until: | - ('No resources found' not in _openshift_containers.stdout) or - ('connect: connection refused' not in _openshift_containers.stderr) or - ('get current server API group list' not in _openshift_containers.stderr) + until: | + ('No resources found' not in _openshift_containers.stdout) and + ('connect: connection refused' not in _openshift_containers.stderr) and + ('get current server API group list' not in _openshift_containers.stderr)
🧹 Nitpick comments (2)
oci/Containerfile (1)
15-15: Confirm base image digest bump and scan for CVEs.Updating the CentOS Stream 9 digest can change libc/openssl and other base libs. Please confirm this digest was intentional, passes image/vuln scanning, and doesn’t break downstream installs. If driven by Renovate, consider annotating with a renovate hint for traceability.
ansible/roles/deploy-crc-cloud/tasks/kubelet.yaml (1)
12-12: Make timeout configurable via a role var.Allows env-specific tuning without future PRs.
- timeout: 600 + timeout: "{{ kube_apiserver_port_wait_timeout | default(600) }}"
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (2)
ansible/roles/deploy-crc-cloud/tasks/kubelet.yaml(1 hunks)oci/Containerfile(1 hunks)
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (3)
- GitHub Check: build (macOS-latest, 1.20)
- GitHub Check: build (ubuntu-latest, 1.20)
- GitHub Check: build-and-push-image
🔇 Additional comments (1)
ansible/roles/deploy-crc-cloud/tasks/kubelet.yaml (1)
12-12: LGTM: Longer kube-apiserver port wait.Increasing timeout to 600s matches the PR goal and should reduce flakiness on slow boots.
3 participants
Sometimes, 300 seconds waiting time for port :6443 to be available is not enough.
Increase the time to 600 seconds.
Summary by CodeRabbit