Alert on high load
As seen on the screenshots below, the droplet for this pipeline (MR !351 (merged)) had very high load (up to 60).
k9s reported 100% CPU usage.
The nc-rabbitmq pod failed to start, and I couldn't debug what was the issue.
Steps to solve this:
-
#438 (closed) -
Give more CPU to the droplets where we test installation of all apps (currently we have 4 CPU cores and 8GB of RAM) -
Monitor droplet resource usage during CI runs (#726)
Edited by Varac