I have an end-to-end scenario to include in performance testing that takes a decent length of time to execute. Between releases of the system, this time needs to be monitored to ensure it hasn't increased.
I am already taking advantage of the very useful pass/fail criteria in Taurus. However, the available criteria are quite granular (at the individual request/response level). I know Taurus does output test duration but this could include other items that I'm not interested in. I also can't see test duration supported as a metric for the pass/fail module in Taurus.
So, any suggestions on how to extract the time to complete the process and then compare this time with a pass/fail criterion? The script itself would be executed within Jenkins for what it's worth.