JIRA 4.3 : Is Clustering or Load Balancing JIRA Possible
This page last changed on Mar 23, 2011 by rosie@atlassian.com.
Does JIRA support clustering for scaling?Not at this time. Currently, splitting JIRA into multiple instances is our recommended means of scaling JIRA — please see Splitting a JIRA instance. Note: If you wish to run JIRA across multiple nodes you will need a license for each node. You can find our discounting policy here. Does JIRA support high-availability?If you are interested in clustering for hardware failover, you can configure a distributed database, load balancer front-end and two separate servers. The primary JIRA instance runs on the main server, while the second server contains an unstarted, free development license instance of JIRA. The load balancer directs 100% of requests to the primary as long as it is responsive, but if the primary server goes down, the load balancer starts the backup instance and directs all requests there until manually reset. There are several problems with this model. Specifically, attachments and indexes will be out of date on the backup server. You may be able to institute a script to update these. |
![]() |
Document generated by Confluence on Mar 27, 2011 18:52 |