[BUG] AksHci upgrade hangs when the upgrade is initiated after adding a physical node to the setup
- Dominant language
- PowerShell
- Stars
- 118
- Forks
- 63
- PR merge metrics
- No merged PRs in 30d
Description
**Describe the bug**
AksHci upgrade hangs if we initiate the upgrade after adding a new physical to an existing AksHci setup. The hang typically looks as shown in the screenshot below.
The CSI controller pod logs has the following error.
**To Reproduce**
Steps to reproduce the behavior:
1. Install AksHci
2. Add a new physical node
3. Perform AksHci upgrade with `Update-AksHci`. The upgrade hangs during this step
**Expected behavior**
Ideally, the upgrade should complete without any issues.
**Mitigation**
1. Drain the node using failover cluster UI as shown below.
Alternatively, you can use the command `Suspend-ClusterNode -Name -Drain` to drain the node.
2. Use `Remove-AksHciNode -nodeName ` to remove the machine from akshci setup
3. Use `Remove-ClusterNode -Name ` to remove the machine from failover cluster
4. Run `Update-AksHci` to trigger the upgrade.
Note: We can remove the node while the upgrade is hanging. The upgrade will proceed.
Contributor guide
Assessment
This issue has not been assessed yet.