My gitlab (CE, 10.5.4) is running inside a VM.
It spawns up to 9 git processes plus 1 gzip and 1 bzip2 process which each take 100% CPU. This causes the host to have a load of up to 200+ until I simply restart gitlab. A few hours later, it’s the same problem.
I’ve checked all the logfiles in /gitlab-rails and the only problem I could find were 5 out of around 100 runners that can’t connect due to a wrong config that will be fixed within the next hours.
Any idea what else I could check or what could be the problem?