harvester: [BUG] Unable to delete the stuck provisioning RKE2 cluster if cluster VM have no IP assign

Describe the bug

When RKE2 cluster VM can’t get IP address and stuck in provisioning status Unable to delete the RKE2 cluster and VM correctly if Harvester VM did not get IP address on it.

  • Delete the RKE2 cluster stuck in Removing
    image

  • Harvester cluster VM keep in Terminating image

  • Harvester cluster VM events show deleted image

To Reproduce Steps to reproduce the behavior:

  1. Import Harvester from Rancher
  2. Provision a 1 node RKE2 cluster in Harvester (v1.22.7+rke2r2)
  3. RKE2 cluster stuck in provisioning status
  4. RKE2 cluster VM created on Harvester but can’t get IP assigned
  5. Delete the RKE2 cluster from Rancher UI
  6. Check can delete RKE2 cluster and Harvester cluster VM

Expected behavior

Should be able to delete RKE2 cluster from Rancher and delete cluster VM from Harvester when VM did not get IP address

Support bundle

supportbundle_58124c47-18ab-424a-9971-5e78b768f145_2022-03-21T13-40-36Z.zip

Rancher logs 0321_regression.log

Environment:

  • Harvester ISO version: v1.0.1-rc1
  • Rancher version: v2.6.4-rc10
  • Underlying Infrastructure (e.g. Baremetal with Dell PowerEdge R630): 2 nodes harvester cluster
  • RKE2 version: v1.22.7+rke2r2

Additional context Add any other context about the problem here.

About this issue

  • Original URL
  • State: closed
  • Created 2 years ago
  • Comments: 16 (11 by maintainers)

Most upvoted comments

Hi @tlehman , I think the issue can still exists on v1.22.8 rke2 version. I guess you are using v2.6-head for the investigation.

According to the dev-2.6 rke-config, I think the v1.22.7-rancher1-2 should be identical with v1.22.7+rke2r2 https://releases.rancher.com/kontainer-driver-metadata/dev-v2.6/data.json

image

image