rancher: Failure to deploy Helm3 App: apiVersion v2 invalid, must be v1

What kind of request is this (question/bug/enhancement/feature request): bug

Steps to reproduce (least amount of steps as possible):

Click Cluster > Apps > Launch
Choose an app from a Helm3 repository and launch

Result: Fails to deploy app, saying wait helm template failed. Error: apiVersion 'v2' is not valid. The value must be "v1"

Other details that may be helpful: The log when deleting the app looks like the following, which looks like it’s trying to use Helm2/Tiller to control the app

2020/03/31 21:06:39 [ERROR] AppController p-p6c5p/ocm [helm-controller] failed with : wait helm template failed. Error: apiVersion 'v2' is not valid. The value must be "v1": exit status 1
---- repeats about once a second until I delete the app -----
2020/03/31 21:06:40 [INFO] Deleting chart using helm version: rancher-helm
[main] 2020/03/31 21:06:40 Starting Tiller v2.16.3-rancher1 (tls=false)
[main] 2020/03/31 21:06:40 GRPC listening on :47294
[main] 2020/03/31 21:06:40 Probes listening on :39590
[main] 2020/03/31 21:06:40 Storage driver is ConfigMap
[main] 2020/03/31 21:06:40 Max history per release is 10
[storage] 2020/03/31 21:06:41 getting release history for "ocm"

Environment information

Rancher version (rancher/rancher/rancher/server image tag or shown bottom left in the UI): v2.4.0
Installation option (single install/HA): single install

Cluster information

Cluster type (Hosted/Infrastructure Provider/Custom/Imported): Custom
Machine type (cloud/VM/metal) and specifications (CPU/memory): VM 4core 4GB ram
Kubernetes version (use kubectl version):

Client Version: version.Info{Major:"1", Minor:"17", GitVersion:"v1.17.3", GitCommit:"06ad960bfd03b39c8310aaf92d1e7c12ce618213", GitTreeState:"clean", BuildDate:"2020-02-11T18:14:22Z", GoVersion:"go1.13.6", Comp
iler:"gc", Platform:"linux/amd64"}
Server Version: version.Info{Major:"1", Minor:"17", GitVersion:"v1.17.3", GitCommit:"06ad960bfd03b39c8310aaf92d1e7c12ce618213", GitTreeState:"clean", BuildDate:"2020-02-11T18:07:13Z", GoVersion:"go1.13.6", Comp
iler:"gc", Platform:"linux/amd64"}

Docker version (use docker version):

Client: Docker Engine - Community
 Version:           19.03.8
 API version:       1.40
 Go version:        go1.12.17
 Git commit:        afacb8b
 Built:             Wed Mar 11 01:27:04 2020
 OS/Arch:           linux/amd64
 Experimental:      false

Server: Docker Engine - Community
 Engine:
  Version:          19.03.8
  API version:      1.40 (minimum version 1.12)
  Go version:       go1.12.17
  Git commit:       afacb8b
  Built:            Wed Mar 11 01:25:42 2020
  OS/Arch:          linux/amd64
  Experimental:     false
 containerd:
  Version:          1.2.13
  GitCommit:        7ad184331fa3e55e52b890ea95e65ba581ae3429
 runc:
  Version:          1.0.0-rc10
  GitCommit:        dc9208a3303feef5b3839f4323d9beb36df0a9dd
 docker-init:
  Version:          0.18.0
  GitCommit:        fec3683

About this issue

Original URL
State: closed
Created 4 years ago
Comments: 24 (7 by maintainers)

Most upvoted comments

After filing this, I tried nuking the catalog (which was labeled Helm3) and recreating it, and it seemed to work.

I did that a couple of times, it seemed to work some of the time, but around half the instances of my apps i try to create in different projects fail with the v1/v2 error again because rancher apparently tries to use helm 2 instead of helm 3.

Are there known cache issues? Can it have to do with re-using names for namespaces / projects / app combinateions where previously a helm 2 app existed? I also made sure that the v3 catalog has a different name than the v2 catalog that existed before, still Rancher tries to deploy apps with helm2 and then fails with the invalid apiVersion v2 error…

What is a good way to help diagnose this issue?

(currently on 2.4.2)

dasMulli on Apr 17, 2020

1 month, no resolution or workaround. So much for an enterprise kubernetes management that can’t even deploy an app without breaking.

swadeshp on Oct 2, 2021

Also exiting here when trying to deploy extrenal-dns from Bitnami helm v3 Rancher 2.5.7

krumware on Jun 17, 2021

@rex yes, same issue trying to deploy the Ghost Blogging platform on a test cluster via the Bitnami helm chart. I can confim the issue still persists. I am using Rancher v2.5.7 with K3S (Kubernetes 1.20.5)

jnfinitym on Apr 10, 2021

@cbron This issue should be reopened, as this problem persists. I am dead in the water right now trying to spin up a new cluster because a critical infrastructure service (bitnami/rabbitmq) refuses to deploy. Tried deploying both Helm 2 and Helm 3 versions, the error is the same. There are other instances of this service running just fine in other clusters, but it refuses to deploy in the new one.

I will not remove and re-add the catalog because so many things deployed to production rely on the catalog and I can’t afford any downtime if removing the catalog causes rancher to freak out. What else is there that we can do? Since apparently upgrading to the most recent version is not even enough to resolve the issue.

rex on Apr 1, 2021

@theAkito the workaround is to create an app with an old version of the chart that was present in the catalog when you created it initially, then update the app to a recent version.

dasMulli on Aug 29, 2020

@brendarearden

Sorry for the misunderstanding. I cannot update my cluster just for fun, so I want to only update, if it really works. I am on Rancher 2.4.5.

theAkito on Aug 28, 2020