Executor Plugin error

[raosharadhi]

Hi Team,
I wanted to explore the plugin part in argo workflows.

• Installed argo workflows (v3.3.8) into the Minikube cluster.
• Added env in workflow controllers and set Argo Executor plugins value as true.
• Cloned the hello-world plugin from the official GitHub link in the documentation and ran these commands.

argo executor-plugin build .
kubectl -n argo apply -f hello-executor-plugin-configmap.yaml

• I Can see the plugin installed in the logs of the workflow controller.

time="2022-07-26T06:33:34.492Z" level=info msg=Plugins executorPlugins=true
time="2022-07-26T06:33:34.520Z" level=info msg="Executor plugin added" name=hello-executor-plugin 

workflow code:

apiVersion: argoproj.io/v1alpha1
kind: Workflow
metadata:
  generateName: hello-
spec:
  entrypoint: main
  templates:
    - name: main
      plugin:
        hello: { }

But when copy and run the same workflow it shows an error like this.

Failed: invalid spec: template 'main' type is unknown

For Screenshots please go here: https://dev.to/sharadhi/argo-workflow-executor-plugins-2imm

Can anyone help me to solve this issue?

The links I referred are: https://github.com/argoproj-labs/argo-workflows-hello-executor-plugin
https://argoproj.github.io/argo-workflows/executor_plugins/

[terrytangyuan]

@alexec Could you take a look?

[alexec]

Can you check that your argo cli is up to date?

[raosharadhi]

Yes This is the version I am using.

[alexec]

I've not been able to reproduce this locally. The workflow runs successfully. This hints to a user error.

Could you try kubectl -n argo create -f hello-example-workflow.yaml?

[raosharadhi]

Now If I ran from UI I get this same error as template unknown and As you suggested If I ran that command. The pod is in a pending state as you can see in second workflow.

I installed Argo workflow using these commands. @alexec Can you please once check Is it because of this?

kubectl apply -n argo -f https://github.com/argoproj/argo-workflows/releases/download/v3.3.8/install.yaml 
kubectl -n argo create rolebinding default-admin --clusterrole=admin --serviceaccount=argo:default 

[maxisam]

I just wanna say I have exactly the same issue. I installed through helm chart v3.3.9. I have RBAC enabled but I am fine to run normal workflow. I can see the plugin got added from the logs as well. I can't see any plugin from the UI

[lijiaonothing]

I use the CLI in 'Kubernetes API Mode (default)', The workflow runs successfully.
when I use CLI in 'Argo Server HTTP1 Mode', it shows an error like this: 'Failed: invalid spec: template 'main' type is unknown'

[alexec]

Can you check the logs of the agent and of the controller to find where that error occurs please?

[raosharadhi]

to check the agent pod, the agent pod is not creating itself. It's deleting by itself.

[alexec]

Can you check the image the agent pod is the same version as the controller.

[raosharadhi]

Hi @alexec , For now, I did cluster install through kustomize. Now plugin I am able to see, but I am getting this error now in agentpod.

time="2022-08-01T11:43:35.308Z" level=warning msg="Non-transient error: workflowtasksets.argoproj.io \"hello-example-s7rsd\" is forbidden: User \"system:serviceaccount:argo:argo\" cannot patch resource \"workflowtasksets/status\" in API group \"argoproj.io\" in the namespace \"argo\""
time="2022-08-01T11:43:35.308Z" level=error msg="TaskSet Patch Failed" error="workflowtasksets.argoproj.io \"hello-example-s7rsd\" is forbidden: User \"system:serviceaccount:argo:argo\" cannot patch resource \"workflowtasksets/status\" in API group \"argoproj.io\" in the namespace \"argo\"" workflow=hello-example-s7rsd

The pod is running successfully 2/2 and I have also added workflowtaskset/status role in the aggregator file.

Can you please look into this?
I used this command

argo submit  hello-example-workflow.yaml --serviceaccount argo -n argo

[chaudharirohit2810]

Has anyone found solution for this? Currently we are facing exact same issue

[sumeet-baghel]

I don't know if my answer is relevant to this specific issue but I observed the same error today while creating a ClusterWorkflowTemplate for DIND use case.

dind-template

apiVersion: argoproj.io/v1alpha1
kind: ClusterWorkflowTemplate
metadata:
  name: dind-template
spec:
  templates:
    - name: docker-sidecar
      sidecars:
      - name: dind
        image: docker:24.0.6-dind # Official docker dind image
        command: [dockerd-entrypoint.sh]
        env:
          - name: DOCKER_TLS_CERTDIR # Docker TLS env config
            value: ""
        securityContext:
          privileged: true # the Docker daemon can only run in a privileged container

        # mirrorVolumeMounts will mount the same volumes specified in the main container
        # to the sidecar (including artifacts), at the same mountPaths. This enables
        # dind daemon to (partially) see the same filesystem as the main container in
        # order to use features such as docker volume binding.
        mirrorVolumeMounts: true

Error while applying

Note

I observed the same error when I applied it directly via kubectl and used it in a workflow

$ argo cluster-template create dind_template.yml
2023/10/06 09:18:29 Failed to create cluster workflow template: dind-template,  rpc error: code = InvalidArgument desc = templates.docker-sidecar template 'docker-sidecar' type is unknown

I was able to resolve the error by adding the container field to the template spec as follows.

apiVersion: argoproj.io/v1alpha1
kind: ClusterWorkflowTemplate
metadata:
  name: dind-template
spec:
  templates:
    - name: docker-sidecar
      container:
        image: "repo/image"
        command: ["/bin/sh", "-c"]
        args: ["ls -l"]
      sidecars:
      - name: dind
        image: docker:24.0.6-dind # Official docker dind image
        command: [dockerd-entrypoint.sh]
        env:
          - name: DOCKER_TLS_CERTDIR # Docker TLS env config
            value: ""
        ...

I'm guessing the container field is required and that's the reason argo is throwing a cryptic error 🤷🏻‍♂️