seed-job-agent container cannot clone the kubernetes-operator repository

[AaronJaegerVA]

Describe the bug

The seed-job-agent container cannot clone the kubernetes-operator repository because it cannot resolve the host github.com; however, testing resolution in the container for github.com seems to work without issue.

bash-5.0$ nslookup github.com
Server:         10.43.0.10
Address:        10.43.0.10:53
Non-authoritative answer:
Name:   github.com
Address: 192.30.255.112
Non-authoritative answer:
. 
.
.
bash-5.0$ git clone https://github.com/jenkinsci/kubernetes-operator.git/
Cloning into 'kubernetes-operator'...
fatal: unable to access 'https://github.com/jenkinsci/kubernetes-operator.git/': Could not resolve host: github.com

Additional information

Kubernetes version: 1.20
Jenkins Operator version: 0.6.2

[AaronJaegerVA]

I know from past experience that alpine behaves strangely when it comes to DNS resolution, so I decided to edit the seed-job-agent deployment to use a different jenkins/inbound-agent image. I switched from jenkins/inbound-agent:alpine to jenkins/inbound-agent:latest-jdk11 and all is well now.

It may make sense to change the default seed-job-agent container image going forward.

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. If this issue is still affecting you, just comment with any updates and we'll keep it open. Thank you for your contributions.

[rafabene]

I'm facing the same issue. Is there any workaround?

[rafabene]

The following command updates the seed-job-agent-master image.

kubectl patch deployment seed-job-agent-master -p '{"spec":{"template":{"spec":{"containers":[{"name":"jnlp","image":"jenkins/inbound-agent:latest-jdk11"}]}}}}'

[Sig00rd]

Thank you for your input. I've managed to reproduce the problem on minikube with latest version of the Operator. It seems to be related to this issue with alpine image: kubernetes/minikube#10830.
inbound-agent:4.9-1 seems all right, so I'll make a quick PR with the image version bump.

[Sig00rd]

With PR bumping the agent image version merged, using the next Operator nightly build (the one scheduled for tomorrow) should fix the problem, I'll update here once I verify it's working correctly and until then, patching seedjob agent deployment as @rafabene suggested should help

[Sig00rd]

I've just tested it, using Operator's nightly build solves the problem. Thank you for reporting this problem and keeping this issue alive!