Uploaded image for project: 'Jira Data Center'
  1. Jira Data Center
  2. JRASERVER-43040

Favourite Filters Issue Count Performance

XMLWordPrintable

    • 6.03
    • 17
    • Severity 2 - Major
    • 54
    • Hide
      Atlassian Update – 8 July 2020

      Hi everyone,

      Thank you for your interest in this issue. We're glad to announce that Favourite Filters Issue Count Performance problem will be addressed in the upcoming Jira Server version 8.11.0.

      We’ve added a new global setting that allows Jira admins to set the max timeout for the Favorite Filters gadget. For large filters, counting and showing the issues might affect Jira performance, so this setting is a fail-safe to make sure everything works right.

      You will be able to set it by going to Administration > System > General configuration. The default value is 5000 ms. If you set it to -1, we won’t count and show the issues at all.

      Kind regards,
      Grażyna Kaszkur
      Product Manager, Jira Server and Data Center

      Show
      Atlassian Update – 8 July 2020 Hi everyone, Thank you for your interest in this issue. We're glad to announce that Favourite Filters Issue Count Performance problem will be addressed in the upcoming Jira Server version 8.11.0. We’ve added a new global setting that allows Jira admins to set the max timeout for the Favorite Filters gadget. For large filters, counting and showing the issues might affect Jira performance, so this setting is a fail-safe to make sure everything works right. You will be able to set it by going to Administration > System > General configuration. The default value is 5000 ms. If you set it to -1, we won’t count and show the issues at all. Kind regards, Grażyna Kaszkur Product Manager, Jira Server and Data Center

      Summary

      JIRA call rest/gadget/1.0/favfilters?showCounts=true returns the total number of issues matching all favourite filters for the user.
      That means it will execute and resolves all JQL (provided search criteria) in those favourites filters.
      Please note that this is not a functional bug, as it works exactly as expected and code has same performance impact as if you run those JQL manually.

      That being said, in some cases those JQL execution could be very expensive and due to frequent gadget refresh can cause memory pressure or even lead to OOM.
      Ticket to update docs and explicitly document possible problems: JRA-64383

      Steps to Reproduce

      1. Create large instance (500k issue, 500 projects)
      2. Create many filters for the specific user
      3. (Optional) Create complex filter using 3rd party plugins
      4. Trigger REST call: rest/gadget/1.0/favfilters?showCounts=true
        • Same call is triggered by Favourite Filters gadget

      Expected Results

      JIRA performs fast

      Actual Results

      JIRA is slow.

      Notes

      • Although this is documented already to be slow and non-performant, users still are using it beyond the control of administrators.
      • With the performance improvements done on 6.3, this is now the slowest operation we see in our instance. We see typical loads times of 8 ~ 10 seconds when Issue Count is turned on.
      • Suggestion: make showCounts=false the default for the gadget.
        • See screenshot

      Workaround

      • Disable favourite count from proxy level
         RewriteCond %\{REQUEST_METHOD} GET
         RewriteCond %\{REQUEST_URI} ^/rest/gadget/1.0/favfilters.*$ [NC]
         RewriteCond %\{QUERY_STRING} ^showCounts=true(.*)$ [NC]
         RewriteRule /rest/gadget/1.0/favfilters /rest/gadget/1.0/favfilters?showCounts=false%1 [R=302,L] 
        
      • Disable for existing gadgets. Option #2. REST api (see below in comments)
      • Disable Favourite Filters gadget
        • Go to: [ Administration > Manage add-ons > System add-ons > Atlassian JIRA - Plugins - Gadgets Plugin]
        • disable Favourite Filters (favourite-filters-gadget)

        1. fav_edit.png
          fav_edit.png
          49 kB
        2. fav_view.png
          fav_view.png
          50 kB

              f62ec3ff2b06 Karol Kulesza (Inactive)
              a38518e05741 David Yu
              Votes:
              16 Vote for this issue
              Watchers:
              29 Start watching this issue

                Created:
                Updated:
                Resolved: