Issue Summary

This is reproducible on Data Center: (yes)

Steps to Reproduce

1. Observe any errors in the log files 
2. Some of the errors could be logged on the warn level
3. UI present that rebalancing finished successfully. 

Example of the error:

[service.lexorank.balance.LexoRankBalancer] Indexing Queue for rebalanced issues overflow, reindex stuck or is taking longer than expected >PT320S 

Expected Results

UI should propagate errors to the UI. Additionally errors shouldn't be logged on warn level.

Actual Results

Rebalancing UI still shows that rebalancing finished successfully.

After refreshing the page only some of the issues being rebalanced. Issues are redistributed between two buckets. 

Workaround

After  fix of JSWSERVER-16057  in Jira 10.2 and later, rebalancing shouldn't stuck for infinite amount of time and you can try to trigger rebalancing once again.