Folks, I've noticed that the perf advisor tests are quite often timing out in the beforeAll clause where we create an M10 cluste# This is not entirely surprising as provisioning can sometimes take a longer time, but it's also a bit wasteful as we're only running them to check that the response contains a couple of headers, which is also something we'd get if we ran the test against an M0. Not sure if there's something smart we can do here - either run this against a long-running cluster or rework the workflow so we run them on merges to main/schedule, but I'm sure the Atlas folks are not super happy about us spending 15-20 minutes on provisioning an M10 to then use it for 2 seconds and tear it down :sweat_smile:

Created with Jira Integration+ for bianca.vianadeaguiar (bianca.vianadeaguiar@mongodb.com)