Hi Comumunity,
this has been a topic for us quite some time and I have a question about it.
What is the maximum allowed value here? On the DB Server, the value is 0, which means unlimited, but in the Jira settings, we seem to have a bottleneck, and this limits us.
<pool-min-size>30</pool-min-size>
<pool-max-size>30</pool-max-size>
In many places in the log, you can see attempts by the jira server to call 31, 32 or 33 connection, which results in rejection and then jira slows down alot.
Has anyone had a similar problem and what is the maximum possible number so that there is no problems for anyone.
We have about 1500 active users.
Best regards
Jurica
This article might be able to offer you some guidance. The answer is, there is no one right answer, a lot of it will depend on your specific usage. If you're seeing errors in your logs indicating a bottleneck that's a good indication you need to increase your max size. The nice thing about these settings is Jira will open and close connections to meet demand according to your pool size. I would also recommend checking out the Database Connection Monitor page in your instance to get an idea of where you're usage is on average. If you're consistently hitting your max during peak hours, increase the max in the dbconfig.
I have seen as a general recommendation to set the max +10 or +25% (whichever is higher) to what your maximum threads are set to in your tomcat configuration. The idea behind this is if your tomcat config allows for the default (200 connections) and you only have 30 available in your pool, there's a potential there for a bottleneck. I believe this was specific for Confluence but I tend to use it for Jira too.
Hope this helps.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.