Operate stuck

Camunda 8 Topics
2

Hi @shira4520,

This is a known issue with Camunda 8.5 self-managed Operate called the “Operate Post-Importer stuck” problem. The error you’re seeing indicates that Operate’s incident post-importer is trying to process incidents whose corresponding process instances are not yet available in Operate’s indices, causing it to retry continuously and potentially blocking further imports.

Quick Fix (Immediate Resolution)

To unblock Operate immediately, add this environment variable to your Operate container:

CAMUNDA_OPERATE_IMPORTER_POSTIMPORTERIGNOREMISSINGDATA="true"

This tells Operate to ignore missing data during post-import, allowing the importer to continue processing new instances instead of getting stuck on the problematic ones. Restart Operate after adding this environment variable.

Root Cause Investigation (Optional)

If you want to investigate the underlying data inconsistency:

  1. Collect the incident IDs from your error logs (like 6755441638815354)

  2. Query Elasticsearch to find these incidents:

    POST <elastic_url>/operate-*,zeebe-record_*/_search
{
  "size": 10000,
  "query": {
    "terms": {
      "key": [6755441638815354]
    }
  }
}

  3. Check the corresponding process instances using the processInstanceKey from the results

  4. For data cleanup, it’s recommended to open a support case to confirm what can safely be removed

Additional Checks

  • Verify your Elasticsearch cluster health and that operate-* and zeebe-* indices are present and green
  • Check Operate logs for similar repeated post-import errors

The environment variable fix should resolve your immediate issue and allow Operate to continue processing new instances normally.

References:

Let me know if you need any clarification or if the issue persists after applying the fix! :slightly_smiling_face: