JIRA's connection pool

Yonatan Leket

Hello,

The error message reported in the Jira logs about the connection pool acquired by a thread is not reporting right numbers,

2019-05-12 13:34:33,311 Caesium-1-1 WARN ServiceRunner [c.a.jira.ofbiz.ConnectionPoolHealthSqlInterceptor] Dangerous use of multiple connections: replaced => count=0; marks=[-1-0]; pool=22/20

As you can see in the attached pic the database Monitor shows a max of 7 connections, 
Although my log file is full of this warning, and it keeps going every couple of seconds. 

I'm not sure what to do, I read that this warning is because my JIRA is suffering from pool exhaustion, but according to the monitor this is not the situation,

Should I stick to the monitor's result or log warnings, What do you guys suggest me to do? 

Andy Heinzer

Hi Yonatan,

This WARN message is a known false positive being generated in the logs.  There is a bug tracking this problem over in JRASERVER-68409.  From the comments of that post:

This is a bug that just affects logging and not performance, so it does cause false positives to be printed in the logs. If you're not seeing any performance issues, then you can safely ignore these errors. However, if you do see any performance issues in the future, please reach out to our support teams at https://support.atlassian.com/contact/ and they will be happy to help!

I hope this helps to clarify this discrepancy.

Andy