Static version of issues for reading purpose

Question

We have a Jira with on average 500 user sessions.  40% of these sessions are from a Jenkins environment - just reading issue data.

We were wondering if it would be possible to cache this issue information on a proxy, such that the Jira doesn't get hit by the request.

Any experience with this type of setup.

francis · Answer

Looking into microcaching in nginx.  It might be an approach.

francis · Answer

It appears that the user account for Jenkins is used by many different teams to do some automation - The engineering team is pretty large, and never trust an engineer with a raspberry pi.  They build the wildest stuff.

One option would be to use data center and reserve one node for the reading activity - but if there is another approach to cache as much as possible downstream, then that would be another option.

Curious if someone built such capability

Mirek · Answer

Isn't this a problem on the Jenkins side? I remember there was a problem with Crowd that Jenkins was "killing" with requests - https://issues.jenkins-ci.org/browse/JENKINS-16703 . Overall I am wondering why it is creating so many sessions. Cannot reuse existing for couple requests together? I remember also one guy had a custom REST API integration with JIRA and also created a lot of sessions only by reading data.. He changed the code and it was a lot better.