I have a kubernetes namespace that I want to leverage for Gitlab runners. I installed the runners following the Helm Chart instructions. The problem I am running into is that when the job container spins up, I get the following ERROR:
Job failed: image pull failed: rpc error: code = Unknown desc = Get https://registry-1.docker.io/v2/: net/http: request canceled while waiting for connection (Client.Timeout exceeded while awaiting headers)
It's trying to connect to the public docker repo but my organizations firewall is blocking it. How would I go about having the instace go to our private repo?
Any help would be greatly appreciated as I've been stuck on this issue for some time now :(
I would presume you'll need to specify a values.yaml to the helm install that points to your mirrored copy of the images it needs. So:
or whatever ones you wish to use for the init and runner: image:
Since you already have the chart deployed, I am fairly certain you can just do an "helm upgrade" that changes only those values:
helm upgrade --set "image=repo.example.com/gitlab/gitlab-runner" \
--set "init.image=repo.example.com/etc-etc" \
[and so forth] \
$releaese_name $chart_name(substituting the release name and the name of the chart as your helm knows it, of course)