Rest API perfomrance issue on production environments

[Daren Porter]

We are experiencing performance issue with Rest Api action, which cause spike in user performance. Do the Atlassian team have any future proposal to mitigate these issues without having to have a dedicated node for rest API actions. Could we have better governance tool or improve balancing of node resources. Giving priority to One node to run API and to balance users when not fully loaded.

[Radek Dostál]

The REST API requests may be written in an inefficient way as well, effectively authenticating against Jira with every single request they make.

We were once looking at some automation bot developed locally by a team that was attempting to authenticate several hundred times a minute. That and the fact it could be trying to chain update hundreds of issues at once.

It might be worth looking at some of the requests and tracking down how extensive/heavy they are, and what their implementation is.

[Nic Brough -Adaptavist-]

What are your users doing to cause these perceived spikes?  How and what are you measuring?