I am using Camunda 8 self managed and getting the following error in APIs on my self-managed operate dashboard. Could anyone guide me as to what could be the issue
{
“timestamp”: “2025-04-28T06:36:17.788+00:00”,
“status”: 500,
“error”: “Internal Server Error”,
“message”: “Elasticsearch exception [type=search_phase_execution_exception, reason=all shards failed]”,
“path”: “/api/incidents/byProcess”
}
Hi @Awais_Sabir Based on the error message: Elasticsearch exception [type=search_phase_execution_exception, reason=all shards failed]
Operate tried to query Elasticsearch (to fetch incidents, from /api/incidents/byProcess) but the query failed on all shards.
Possible causes:
- Elasticsearch is down or not reachable from Operate.
- Index corruption or missing index that Operate expects.
- Elasticsearch cluster health is red (some shards are unassigned or failed).
- Mapping or schema mismatch in the Operate-related Elasticsearch indices.
- Authentication issues (if you have security enabled between Operate and Elasticsearch).
- Resource problems — e.g., Elasticsearch is out of disk, memory, or CPU.
Check Elasticsearch cluster health:
curl -X GET "http://<your-elasticsearch-host>:9200/_cluster/health?pretty"
If it shows status: red, shards are failing.
- Check Elasticsearch logs (very important):
Look for errors like shard failure, disk full, circuit breaker, index missing, etc.
- Check the indices used by Operate:
Operate stores data in Elasticsearch indices like:
operate-*
zeebe-*
You can list indices:
curl -X GET "http://<your-elasticsearch-host>:9200/_cat/indices?v"
- If you recently upgraded Camunda or reinstalled, Operate might expect indices that don’t exist.
- Or if Elasticsearch data was wiped/reset incorrectly.
- Check disk usage on Elasticsearch nodes.
- Check memory/heap usage.