Can I specify the private image for step-move-all-results-to-tekton-home/step-copy-artifacts tasks

[yichenwan]

/kind question

Question:
[You can ask any question about this project.]

Hi I currenty run the pipeline on k8s clusters, all the images the pod pull is from the private repostiry, but after I run my pipeline, I met the problems as beflow:

  Warning  Failed     2m29s                kubelet            Failed to pull image "busybox:1.34.1": rpc error: code = Unknown desc = failed to pull and unpack image "docker.io/library/busybox:1.34.1": failed to resolve reference "docker.io/library/busybox:1.34.1": failed to do request: Head "https://registry-1.docker.io/v2/library/busybox/manifests/1.34.1": dial tcp 3.94.224.37:443: i/o timeout
  Normal   Pulling    89s (x2 over 3m29s)  kubelet            Pulling image "busybox:1.34.1"
  Warning  Failed     89s                  kubelet            Failed to pull image "minio/mc:RELEASE.2020-11-25T23-04-07Z": rpc error: code = Unknown desc = failed to pull and unpack image "docker.io/minio/mc:RELEASE.2020-11-25T23-04-07Z": failed to resolve reference "docker.io/minio/mc:RELEASE.2020-11-25T23-04-07Z": failed to do request: Head "https://registry-1.docker.io/v2/minio/mc/manifests/RELEASE.2020-11-25T23-04-07Z": dial tcp 3.94.224.37:443: i/o timeout
  Warning  Failed     89s                  kubelet            Error: ErrImagePull
  Warning  Failed     29s (x2 over 2m29s)  kubelet            Error: ErrImagePull
  Normal   Pulling    29s (x2 over 2m29s)  kubelet            Pulling image "minio/mc:RELEASE.2020-11-25T23-04-07Z"
  Warning  Failed     29s                  kubelet            Failed to pull image "busybox:1.34.1": rpc error: code = Unknown desc = failed to pull and unpack image "docker.io/library/busybox:1.34.1": failed to resolve reference "docker.io/library/busybox:1.34.1": failed to do request: Head "https://registry-1.docker.io/v2/library/busybox/manifests/1.34.1": dial tcp 98.85.153.80:443: i/o timeout

It seems that the image is hard code from the configuration server deployed by one of pods:

$
[u00059002@f12ak8swg176 mlops]$ kubectl get pod -n tekton-pipelines
I0422 17:10:24.177581 3552835 request.go:665] Waited for 1.179529203s due to client-side throttling, not priority and fairness, request: GET:https://10.11.233.179:8443/apis/project.cattle.io/v3?timeout=32s
NAME                                           READY   STATUS    RESTARTS   AGE
tekton-pipelines-controller-5497954b74-6jgjz   1/1     Running   0          85m
tekton-pipelines-webhook-55757d77df-kbx5j      1/1     Running   0          85m

And I found the hard code place:

kfp-tekton/backend/src/apiserver/template/tekton_template.go

Line 353 in 0b89419

Image: common.GetMoveResultsImage(),

The tekton-pipeline yaml I installed is this version: https://github.com/tektoncd/pipeline/releases/tag/v0.39.0

How can I fix it or avoid it?