Failover to internal search when Elasticsearch fails
Problem to solve
If Elasticsearch goes down then GitLab will throw a 500 error on search.
Failover to the internal search when a connection cannot be made to Elasticsearch.
What does success look like, and how can we measure that?
- Stop elasticsearch service
- Search issues
- Confirm that issues are returned without any error
Links / references
(Issues are for internal use only)