Skip to content

Conversation

@jeremyeder
Copy link
Collaborator

…oads

ClickHouse scaled 4x (1Gi→4Gi request, 2Gi→8Gi limit) to support increased trace volume. Zookeeper scaled proportionally (256Mi→1Gi, 512Mi→2Gi) to maintain coordination at new scale.

Evidence of memory pressure:

  • 74 MEMORY_LIMIT_EXCEEDED errors in 15 seconds during merge operations
  • ClickHouse auto-lowered all caches to 1GiB due to "limited RAM"
  • max_server_memory_usage constrained to 1.80 GiB (90% of 2Gi limit)
  • Actual usage during merges: 1.61-1.95 GiB, consistently exceeding limits

Supporting evidence: image

…oads

ClickHouse scaled 4x (1Gi→4Gi request, 2Gi→8Gi limit) to support increased
trace volume. Zookeeper scaled proportionally (256Mi→1Gi, 512Mi→2Gi) to
maintain coordination at new scale.

Evidence of memory pressure:
- 74 MEMORY_LIMIT_EXCEEDED errors in 15 seconds during merge operations
- ClickHouse auto-lowered all caches to 1GiB due to "limited RAM"
- max_server_memory_usage constrained to 1.80 GiB (90% of 2Gi limit)
- Actual usage during merges: 1.61-1.95 GiB, consistently exceeding limits

Supporting evidence: /tmp/clickhouse-memory-evidence.md

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
@jeremyeder
Copy link
Collaborator Author

  containers:
    - resources:
        limits:
          cpu: '1'
          memory: 8Gi
        requests:
          cpu: 500m
          memory: 4Gi

@bobbravo2 bobbravo2 added this to the v0.0.17 milestone Jan 13, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants