We are on JIRA 6.1.7 and over a range of browsers (IE9, Chrome, Firefox) experiencing extremely slow issue searching. Searching for issues in the Navigator hang continuously and our jira-slow-queries.log is full of log messages. Nowhere have I been able to see how we can overcome these slow queries.
Our Environment is virtualised on VMWare.
4 CPU Cores (utilization running at around the 60% mark)
4GB RAM (JVM settings 1024 PermGen, with 2GB Min & Max)
CPU and Memory analysis from Java Melody do not show any particular strain on either of these.
The Disk Speed Tests came back as Good or Excellent using the JIRA support utility in the Knowledgebase article
The Database is MySQL and resides on the same server as JIRA.
The Database connection pools settings appear to be fine
We have implemented the Browse Projects permission for "Anyone"
Our JIRA has approx 70k issues, 160 projects, 250 custom fields, and any plugins we use (about 15) are the latest compatible ones for the version of JIRA we are running. It is connected to Active Directory, and we have the AD connector filter to return approx 1500 users and 12 groups.
Has anyone out there got any ideas on why our JIRA is running so slowly?
https://confluence.atlassian.com/display/JIRA/JIRA+Requirements#JIRARequirements-3.ApplicationServer Since you are probably running 64 bit you should experiment with increasing your VM memory. Running with 1G is definitely too small. You should be able to memory usage under admin/system screen.
John and Norman, thank you kindly for your responses. I plan to upgrade our instance to JIRA 6.3.6 this weekend to hopefully see some improvements.
The default behaviour "Search for issues" to bring up the Issue Navigator is a real annoyance for our user base. It first loads all the issues in our instance, before allowing users to enter the JQL query they actually want to run. Everyone complains about it. Can this be changed?
To clarify, do you not believe the following VM setttings are enough for a JIRA instance of our size -
MaxPermSize=1024m -Xms2048m -Xmx2048m
And finally, could reducing the default user session limit down from 5hrs help with performance?
JIRA does remember the last search, so if users select "current search" instead of "Search for issues" this would avoid the full search for the most part. We also added the "Favorite Filters" gadget to the dashboard, so users can avoid pulling down the menu to click on their saved searches(filters).
I am fairly confident that MaxPermSize is too small, but I usually need to see the JIRA memory usage to make an educated guess since memory usage depends on many things. This also means that most likely you should also add memory to your physical machine. With JIRA and MySQL on the same machine you should be memory starved.with only 4GB.
Click on the diagnosis link to see the information I am pointing you to.
Conor, troubleshoot performance issues is always a challenge since it envolves multiple variables. I recommend to take a look on the following documentation which provide some guidelines to troubleshoot this sort of issue.
Taking thread/heap dumps and enabling the garbage collector logs always provide some pointers of what might be wrong with your instance.
Hope it helps!
Atlassian Support Engineer
In the past, Portfolio for Jira required a high degree of detail–foresight that was unrealistic for many businesses to have–in order to produce a reliable long-term roadmap. We're tur...
Connect with like-minded Atlassian users at free events near you!Find a group
Connect with like-minded Atlassian users at free events near you!
Unfortunately there are no AUG chapters near you at the moment.Start an AUG
You're one step closer to meeting fellow Atlassian users at your local meet up. Learn more about AUGs