Failed to start Portworx

Portworx on Kubernetes
#1

Hands up - I deleted all the AWS EBS volumes that said ‘Do not delete’. I created a new spec on Portworx and installed hoping it would create new volumes – it didnt.

I am receiving this error after the reinstall on my PortWorx + PortWorx API pods on EKS k8s:

Failed to start Portworx: error loading node identity: Cause: ProviderInternal Error: InvalidVolume.NotFound: The volume 'vol-01025c46aefe8f458' does not exist.
       status code: 400, request id: debc0c11-acd3-438c-8646-f01e6ab7fb82

Describe pods:

kube-system   portworx-api-5s9bw                         0/1     Running   0          44m
kube-system   portworx-api-c47cp                         0/1     Running   0          44m 
kube-system   portworx-api-pzd4q                         0/1     Running   0          44m
kube-system   portworx-m24qj                             0/1     Running   2          44m

Anyone point me in the right direction so I can cause more carnage? What is happening?

Much appreciated.

#2

Are you trying to re-install Portworx? If that is the case follow this doc to uninstall other components that were created in Kubernetes by Portworx

Specifically in your case, I suspect that a Kubernetes config map that Portworx uses to track the EBS volumes is not deleted from the previous install.

1 Like