Executing expensive queries that can bog down jira

Phet Phosy August 3, 2016

What debug log file can we enable to pinpoint someone running extensive reports, expensive JQL, etc .. that can slow JIRA performance down to a crawl... aside from looking at slow-queries.log

2 answers

0 votes
Phet Phosy August 4, 2016

Yeah we have a ton of folks who uses agile boards and at a glance, some of the filters(queries) are crazy..times that with other variables, such as plugins, etc and its definitely going to hit at some point.. though  our stack is pretty top notch running on SSDs , dedicated physical DB running MySql, network, ram, etc.. yada yada... 

Would be nice to find a way to search for a crazy filter and/or query folks uses on their dashboards and/or agile/scrum boards, etc without having to go to each 1by1 and also page/alert you when a query takes longer than a minute to execute.

Tarun Sapra
Community Leader
Community Leader
Community Leaders are connectors, ambassadors, and mentors. On the online community, they serve as thought leaders, product experts, and moderators.
August 4, 2016

You can try this plugin https://marketplace.atlassian.com/plugins/net.bull.javamelody/cloud/overview  and then use "was" operator in JQL and see the kind of reports the plugin generates.

Phet Phosy August 5, 2016

Great tool Tarun, thanks –installed this on our dev environment, service is very light weight, seems to keep data for a very long time too.  Hopefully its light weight on space too smile

Will give this a shot on the dev environment using an aggressive jql with the "was" and see what I can break along the process.

 

 

0 votes
Tarun Sapra
Community Leader
Community Leader
Community Leaders are connectors, ambassadors, and mentors. On the online community, they serve as thought leaders, product experts, and moderators.
August 3, 2016

The worst thing you can do to easily bog down a large JIRA instance is to use "WAS" operator in the JQL. But it's just the tip of iceberg, there are other things like low memory, not using SSD drives, slow network, too many issues in rapid-boards, using poorly designed plugin functions in JQL is a major cause of slow JIRA.

Suggest an answer

Log in or Sign up to answer