We are at the point where JIRA is requiring high availability and looking at load balancing. Having read some of the JIRA documentation it seems that the only way this will be possible is to use the JIRA datacenter (I believe you need version 7 JIRA?) and it's not something that is easily setup if this is not used due to the lucine index? I am really just looking to see if this is indeed true or have others successfully setup load balancing without using datacenter?
There have been attempts in the past, but nothing I'd hold up as even a good step towards a success.
You need to move to Data Centre if you want it to work and be supported. It does work for later versions of JIRA 6, but I'd strongly recommend using JIRA 7 for it.
Is this mainly because of the lucine index issues or what are the issues generally with trying to do this? Thanks
From memory, there's three massive problems. One is the index - shared storage simply isn't fast enough to make it usable or even stable. Another one is the session information. I can't remember the third one with any certainty, but I think it was issues with caching and memory states being inconsistent between nodes, which means data can go missing.
We are looking at evaluating Data Centre - even though we won't need datacenter in our DEV/UAT environments do we install it here? Just trying to understand what DC is - is it just an extension of JIRA software that allows clustering? So if we don't use it for clustering in Dev/Test then we set it up just like JIRA Sofware? Documentation just not so clear as it is for JIRA software, hence the confusion.
It looks like you're new here. Sign in or register to get started.
