Uh oh, an old issue seems to have been reincarnated in vSphere 5.5 U2 which may cause vMotions of guests on NFS storage to fail with the error: remote host IP_Address failed with status Busy
This issue was first evident some time ago where it affected V4.1 and was subsequently fixed (albeit temporarily) with ESXi 4.1 u2 +.
See the original KB here with the workaround for those who can’t/couldn’t upgrade; These things happen, we just have to deal with them.
- Launch the vSphere Client and log in to your vCenter Server.
- Select the source ESX host and then click the Configuration tab.
- Click Software > Advanced Settings > Migrate.
- Under the Migrate options, locate the line containing Migrate.VMotionResolveSwapType. By default, it is set to 1.
- Change the value to 0.
- Click OK.
- Repeat Steps 2 to 6 for the destination host.
….but when you have a cluster of many hosts, that’s going to take waaaaaay tooooo long. So, of course, we’re going to speed that up with PowerCLI !
Check your hosts (filter the host list via cluster/datacenter; (I’m filtering to a DEV cluster with 4 hosts)
|
1 |
Get-Cluster "ClusterName" | Get-VMHost | Get-AdvancedSetting -Name Migrate.VMotionResolveSwapType |Sort-Object Entity | ft Entity, Name, Value -Autosize |
Change them;
|
1 |
Get-Cluster "ClusterName" | Get-VMHost | Get-AdvancedSetting -Name Migrate.VMotionResolveSwapType |Set-AdvancedSetting -Value 1 -Confirm=$false |
Verify change
|
1 |
Get-Cluster "ClusterName" | Get-VMHost | Get-AdvancedSetting -Name Migrate.VMotionResolveSwapType |Sort-Object Entity | ft Entity, Name, Value -Autosize |
Keep an eye on the KB, no doubt a patch will be forthcoming.
Job Done. Don’t be CLI shy.