Skipping Deploy due to error: apply: kubectl apply

[cliffburdick]

Expected behavior

Previous job is deleted

Actual behavior

Error on kubectl apply

Information

Hi, I'm trying to understand if this workflow is even possible, because I'm getting an error when deploying. I'm using skaffold dev to deploy a kind 'Job' from the manifest, that runs to completion with the container saying 'Completed' in kubectl get pods. Once it completes, the container exits and skaffold dev sits there. This is all expected. From another window I trigger an update to the source that skaffold dev notices, and it attempts a redploy. This fails with the following error:

Build complete in 13.958695629s
Starting deploy...
kubectl client version: 1.11
The Job "test-container" is invalid: spec.template: Invalid value: core.PodTemplateSpec{ObjectMeta:v1.ObjectMeta{Name:"", GenerateName:"", Namespace:"", SelfLink:"", UID:"", ResourceVersion:"", Generation:0, CreationTimestamp:v1.Time{Time:time.Time{wall:0x0, ext:0, loc:(*time.Location)(nil)}}, DeletionTimestamp:(*v1.Time)(nil), DeletionGracePeriodSeconds:(*int64)(nil), Labels:map[string]string{"controller-uid":"eef54fec-c6b2-11e8-a478-246e96111b94", "job-name":"triad"}, Annotations:map[string]string(nil), OwnerReferences:[]v1.OwnerReference(nil), Initializers:(*v1.Initializers)(nil), Finalizers:[]string(nil), ClusterName:""}, Spec:core.PodSpec{Volumes:[]core.Volume(nil), InitContainers:[]core.Container(nil), Containers:[]core.Container{core.Container{Name:"test-container", Image:"myrepo/app/uncommitted:c907e53-dirty-7fc2439", Command:[]string{"/src/app"}, Args:[]string{"test"}, WorkingDir:"", Ports:[]core.ContainerPort(nil), EnvFrom:[]core.EnvFromSource(nil), Env:[]core.EnvVar(nil), Resources:core.ResourceRequirements{Limits:core.ResourceList{"nvidia.com/gpu":resource.Quantity{i:resource.int64Amount{value:1, scale:0}, d:resource.infDecAmount{Dec:(*inf.Dec)(nil)}, s:"1", Format:"DecimalSI"}}, Requests:core.ResourceList(nil)}, VolumeMounts:[]core.VolumeMount(nil), VolumeDevices:[]core.VolumeDevice(nil), LivenessProbe:(*core.Probe)(nil), ReadinessProbe:(*core.Probe)(nil), Lifecycle:(*core.Lifecycle)(nil), TerminationMessagePath:"/dev/termination-log", TerminationMessagePolicy:"File", ImagePullPolicy:"IfNotPresent", SecurityContext:(*core.SecurityContext)(nil), Stdin:false, StdinOnce:false, TTY:true}}, RestartPolicy:"Never", TerminationGracePeriodSeconds:(*int64)(0xc428003960), ActiveDeadlineSeconds:(*int64)(nil), DNSPolicy:"ClusterFirst", NodeSelector:map[string]string(nil), ServiceAccountName:"", AutomountServiceAccountToken:(*bool)(nil), NodeName:"", SecurityContext:(*core.PodSecurityContext)(0xc425cd0460), ImagePullSecrets:[]core.LocalObjectReference{core.LocalObjectReference{Name:"regcred"}}, Hostname:"", Subdomain:"", Affinity:(*core.Affinity)(nil), SchedulerName:"default-scheduler", Tolerations:[]core.Toleration(nil), HostAliases:[]core.HostAlias(nil), PriorityClassName:"", Priority:(*int32)(nil), DNSConfig:(*core.PodDNSConfig)(nil), ReadinessGates:[]core.PodReadinessGate(nil)}}: field is immutable
WARN[0132] Skipping Deploy due to error: apply: kubectl apply: exit status 1
Watching for changes...
Starting build...

• Skaffold version: v0.15.1
• Operating system: Ubuntu 18.04
• Contents of skaffold.yaml:

apiVersion: skaffold/v1alpha3
build:
  artifacts:
  - docker:
      buildArgs: {}
      dockerfilePath: Dockerfile
    imageName: myrepo/app/uncommitted
    workspace: .
  tagPolicy:
    gitCommit: {}
deploy:
  kubectl:
    manifests:
    - test_manifest*
kind: Config

Steps to reproduce the behavior

1. Create kubernetes Job with restartPolicy Never
2. Run skaffold dev

[r2d4]

This will be fixed by #940. You'll need kubectl v1.12.0 or greater for it to work though!

[cliffburdick]

@r2d4 thanks!

[cliffburdick]

@r2d4 do users need to supply the --force flag manually in skaffold.yaml before this works? I upgraded kubectl to 1.12, but the issue still occurs.

[priyawadhwa]

@cliffburdick once that PR is merged, skaffold will automatically apply the --force flag. You'll just need to update skaffold upon the next release so that this change is included.