cameleer/cameleer-server
1
0
Fork 0
Code Issues 41 Pull Requests Actions Packages Projects Releases Wiki Activity

Add K8s liveness/readiness probes for server and ClickHouse #35

New Issue
Closed
opened 2026-03-12 21:23:23 +01:00 by claude · 2 comments
claude commented 2026-03-12 21:23:23 +01:00
Owner
Copy Link

Problem

Neither the server Deployment nor the ClickHouse StatefulSet have liveness or readiness probes configured. If either process hangs or becomes unresponsive, K8s won't detect it or auto-restart the pod.

Solution

Server (deploy/server.yaml)

  • Readiness probe: GET /actuator/health (Spring Boot Actuator, likely already on classpath) or implement GET /api/v1/health (issue #30)
  • Liveness probe: same endpoint with longer thresholds

ClickHouse (deploy/clickhouse.yaml)

  • Readiness probe: GET /ping on port 8123 (built-in ClickHouse health endpoint, returns "Ok.\n")
  • Liveness probe: same endpoint

Impact

Auto-recovery from hangs/crashes without manual intervention.

## Problem Neither the server Deployment nor the ClickHouse StatefulSet have liveness or readiness probes configured. If either process hangs or becomes unresponsive, K8s won't detect it or auto-restart the pod. ## Solution ### Server (`deploy/server.yaml`) - Readiness probe: `GET /actuator/health` (Spring Boot Actuator, likely already on classpath) or implement `GET /api/v1/health` (issue #30) - Liveness probe: same endpoint with longer thresholds ### ClickHouse (`deploy/clickhouse.yaml`) - Readiness probe: `GET /ping` on port 8123 (built-in ClickHouse health endpoint, returns "Ok.\n") - Liveness probe: same endpoint ## Impact Auto-recovery from hangs/crashes without manual intervention.
claude commented 2026-03-12 22:09:59 +01:00
Author
Owner
Copy Link

Note: ClickHouse is now exposed externally via NodePort (30123 HTTP, 30900 native). Liveness/readiness probes are even more important now that the server depends on ClickHouse being ready before it can initialize its schema. Without a readiness probe on ClickHouse, the server may start before ClickHouse is ready, causing schema init to fail.

Note: ClickHouse is now exposed externally via NodePort (30123 HTTP, 30900 native). Liveness/readiness probes are even more important now that the server depends on ClickHouse being ready before it can initialize its schema. Without a readiness probe on ClickHouse, the server may start before ClickHouse is ready, causing schema init to fail.
claude commented 2026-03-13 10:57:50 +01:00
Author
Owner
Copy Link

Implemented: Added liveness and readiness probes to both deployments.

  • Server: httpGet /api/v1/health:8081 (liveness: 30s initial delay, readiness: 10s)
  • ClickHouse: httpGet /ping:8123 (liveness: 15s initial delay, readiness: 5s)
Implemented: Added liveness and readiness probes to both deployments. - **Server**: `httpGet /api/v1/health:8081` (liveness: 30s initial delay, readiness: 10s) - **ClickHouse**: `httpGet /ping:8123` (liveness: 15s initial delay, readiness: 5s)
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
accessibility
Accessibility (a11y) improvements
 bug
Something isn't working correctly
 cleanup
Code cleanup, dead code removal, refactoring
 epic
Epic: groups related issues
 feature
New functionality
 pmf
Product-market fit: critical for first market offer
 route-diagram
Route diagram page and execution overlay
 security
Authentication, authorization, RBAC
 ui
Frontend / React UI
 ux
UX/usability improvements
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
claude
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: cameleer/cameleer-server#35
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?