Operations Stuck in Camunda

Camunda 8 Topics Discussion & Questions
4

Perfect! Thank you for the details. With Camunda 8.7.12 Self-Managed on Docker, this issue is quite common and there are specific troubleshooting steps we can take without losing your historical data.

Immediate Diagnostic Steps:

1. Check Elasticsearch Health

Run these commands to check your Elasticsearch status:

# Check cluster health
curl -X GET "http://localhost:9200/_cluster/health?pretty"

# List all indices to ensure required ones exist
curl -X GET "http://localhost:9200/_cat/indices?v"

Look for:

  • Cluster status should be green or yellow (not red)
  • Required indices like operate-*, zeebe-* should be present
  • No failed shards

2. Review Container Logs

Check logs for errors in this order:

# Elasticsearch logs
docker logs <elasticsearch-container-name> --tail=100

# Zeebe logs (look for exporter errors)
docker logs <zeebe-container-name> --tail=100

# Operate logs (look for import/query failures)
docker logs <operate-container-name> --tail=100

3. Common Error Patterns to Look For:

In Operate logs:

Exception occurred when performing post import for partition
Process instance is not yet imported for incident processing
Elasticsearch exception [type=search_phase_execution_exception, reason=all shards failed]

In Zeebe logs:

Failed to export record
Elasticsearch exporter error

Quick Recovery Steps (Data-Safe):

Step 1: Restart Services (Ordered)

# Stop services
docker stop <operate-container-name>
docker stop <zeebe-container-name>

# Start services (Zeebe first, then Operate)
docker start <zeebe-container-name>
# Wait 30 seconds
docker start <operate-container-name>

Step 2: Check Resource Usage

# Check disk space
df -h

# Check container resource usage
docker stats

Ensure you have:

  • At least 10% free disk space
  • Sufficient memory for Elasticsearch (recommended: 4GB+)

Advanced Troubleshooting:

If the basic steps don’t work, we can:

  1. Check specific indices integrity
  2. Verify Zeebe-Elasticsearch synchronization
  3. Look for specific data inconsistencies

Questions for Better Diagnosis:

  1. When did this issue first start occurring? (After an update, restart, or specific operation?)
  2. Are you seeing any specific error messages in the UI or logs?
  3. How much disk space is available on your system?
  4. What’s your Docker Compose configuration? (especially Elasticsearch memory settings)

Could you run the health check commands above and share the results? Also, if you can share any relevant log entries from when operations get stuck, that would help pinpoint the exact cause.

This approach should help us identify and fix the root cause without losing your process history! :wrench:

References: