/status should indicate when indexes are broken on a node

Atlassian Update – 15 September 2021 Hi everyone, We’re glad to announce that, starting with Jira 8.19.1, this bug will be fixed. The status health check will now reflect the indexing state. If you were directed to this page for help, you will find more information below. If the node just started and the index isn’t ready yet or it became inconsistent while the node was running, the status health check will respond with 503 HTTP code and { "state":"MAINTENANCE" }. This will tell the load balancer not to route traffic to that node (Data Center only). You can also expect this message to appear in the logs:   INFO [c.a.jira.servlet.ApplicationStateResolverImpl] Checking index consistency. Time taken: 16.16 ms WARN [c.a.jira.servlet.ApplicationStateResolverImpl] The issue index is inconsistent. This node will report its status as MAINTENANCE. You will find information on how to resolve this problem here: https: //jira.atlassian.com/browse/JRASERVER-66970 What to do if status health check returns 503 The first thing to do is to simply wait; more often than not this is a temporary state. Jira will try to recover the index from a snapshot and, if successful, the health check will start returning 200. If the status doesn’t change for a long period (duration depends on the size and configuration of a given instance), you should intervene. Please read Lucene Index Synchronization in Jira Data Center to understand how the index is distributed between Jira nodes. To resolve this issue, follow the steps from Index Replication Jira Data Center Troubleshooting knowledge base article. If none of this helps, you can bring back the old behavior. How to bring back the old behavior Set the system property -Dcom.atlassian.jira.status.index.check=false . Follow these instructions: Setting properties and options on startup . After a restart, Jira will behave as before: the health check won’t reflect the indexing status.