Recover reconciles from recoverAbortedReconcileOnPodReady ignores readiness of service selector labels
Maintainers usually reply within 1 day
Nobody has claimed this yet.
Assessment
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Newbie friendliness
- 35/100
- Issue type
- Bug
- Clarity
- Mostly clear
- Activity status
- Active
- Tech stack
- go, kubernetes
- Domain
- backend, infrastructure
Research direction
Start at the recoverAbortedReconcileOnPodReady path, which decides whether to restart a reconcile from pod readiness alone. Then check how the clickhouse.altinity.com/ready pod label is set and how the chop-generated service selector depends on it. Reproduce the 1-shard, 2-replica sequence from the issue. Done means a recovered reconcile no longer removes ready replicas from the service, and a test covers the aborted-then-ready case.
Written by the indexing model from the issue text.
Description
When an aborted reconcile triggers a recovery reconcile via recoverAbortedReconcileOnPodReady, it uses the pod readiness state, but does not check if the pod labels have clickhouse.altinity.com/ready=Yes.
The result is the chop-generated service might get into a state without any replicas selected despite at least one replica being available to serve queries.
Steps on what occurred in a 1-shard, 2-replica cluster:
- reconcile triggered
- 0-0 starts and completes reconciles
- 0-1 starts reconcile, label set
clickhouse.altinity.com/ready=Noand 0-1 is excluded from service. - 0-1 times out and reconcile is aborted
- 0-1 pod became ready, but still remains excluded from service
- Since 0-1 pod is ready, reconcile recover triggers and starts with 0-0 first making both replicas removed from svc
Operator version: 0.27.4
- Dominant language
- Go
- Stars
- 2.6k
- Forks
- 579
- Avg merge
- 5d 22h
- Merged PRs (30d)
- 5
Getting set up
- Ships a Dockerfile or Docker Compose file
- Has a pull request template
- Read the contributing guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
More from Altinity/clickhouse-operator
-
Counter metrics are missing the conventional _total suffixPossibly taken A pull request linked to this issue is open or already merged. Open
Difficulty 2/5 1-3 hours Newbie friendliness 72/100
Altinity/clickhouse-operator#2093 · 1 comment ·
Maintainers usually reply within 1 day
-
Difficulty 5/5 Over a week Newbie friendliness 12/100
Altinity/clickhouse-operator#2107 ·
Maintainers usually reply within 1 day
-
Difficulty 5/5 Over a week Newbie friendliness 18/100
Altinity/clickhouse-operator#2105 ·
Maintainers usually reply within 1 day
-
Difficulty 4/5 3-5 days Newbie friendliness 35/100
Altinity/clickhouse-operator#2098 ·
Maintainers usually reply within 1 day
-
Difficulty 4/5 3-5 days Newbie friendliness 35/100
Altinity/clickhouse-operator#2096 ·
Maintainers usually reply within 1 day
All issues in Altinity/clickhouse-operator
Similar issues
-
triage
Difficulty 2/5 1-3 hours Newbie friendliness 86/100
wakatime/wakatime-cli#1649 ·
Maintainers usually reply within 1 day
-
Difficulty 2/5 1-3 hours Newbie friendliness 68/100
danielgtaylor/huma#1135 ·
Maintainers usually reply within 1 day
-
needs-triage
Difficulty 2/5 1-3 hours Newbie friendliness 68/100
stacklok/toolhive-catalog#1734 ·
Maintainers usually reply within 1 day
-
documentation priority:p1
Difficulty 2/5 1-3 hours Newbie friendliness 78/100
yeixio/toskar-core#697 ·
Maintainers usually reply within 1 day
-
Difficulty 2/5 1-3 hours Newbie friendliness 68/100
keyxmakerx/Chronicle#1532 ·
Maintainers usually reply within 1 day