3-node cluster of HP DL380 Gen 8's with H220 passthrough controller, 1 SSD and 2 magnetic disks each. vCenter 6 & vSphere 6. HP Support would like me to replace a magnetic drive based on some errors in the iLO (presumably unrelated to this question). The hosts all seem to be performing well.
I've put the host into maintenance mode. http://blogs.vmware.com/storage/2014/12/02/vmware-virtual-san-operations-replacing-disk-devices/ describes removing a disk from VSAN before physically replacing it. That simply doesn't seem to be an option for me, no delete button. Am I missing something or have an problematic configuration? I haven't tried ESXCLI or RVC yet, as this concerned me.
Thanks,
Update: Found that the VSAN cluster disk claiming mode was set to Automatic. Needs to be Manual to see the disk delete button.
However I still have a couple issues. The LED button does not turn the disk LED's on, so I'm not entirely sure what drive I'm working on. And attempting to remove the disk in the GUI says:
A general system error occurred: Failed to evacuate data for disk uuid <uuid> with error: Out of resources to complete the operation
..presumably because this is a 3-node cluster. Calling VMware.
Hi,
Having a similar issue here.
I have a 3 host setup where each host consisted of a single disk group. When I try to remove a disk group I am prompted to move the data to another disk group to prevent data loss. Refer to the screen shot below.
When I click Yes, the procedure is initiated but fails with the following.
Any ideas?
Thank you in advance.
Removing a disk group in a 3 node cluster with only 3 disk groups is not possible.
vSan requires at least 3 disk groups spread over at least 3 hosts.
You could try to add a new disk group before removing the old one.
I think you could remove the disk group if you set the claim to manual and then all VMs supplied a policy with FTT = 0. That way all VMs would be in compliance and you could remove the disk group. Of course you would be working without a net. Thank you, Zach.
Please try below
1. Log in to each ESXi host's console, see Using ESXi Shell in ESXi 5.x and put the host in maintenance mode and select the full data migration option.
2. To obtain information about the VSAN storage, run this command:
# esxcli vsan storage list & # esxcli vsan storage automode set --enabled false
3.To remove the specific SSD disk used in VSAN configuration, run this command:
# esxcli vsan storage remove -s [SSD-DEVICE-ID] & # esxcli vsan cluster get
4. To disconnect the host from the VSAN cluster, run this command:
# esxcli vsan cluster leave & # esxcli vsan cluster get
From the Hosts tab of the cluster, ensure no ESXi host is part of the VSAN cluster. After confirmation, right-click on the cluster and then click Remove.
Please try below
1. Log in to each ESXi host's console, see Using ESXi Shell in ESXi 5.x and put the host in maintenance mode and select the full data migration option.
2. To obtain information about the VSAN storage, run this command:
# esxcli vsan storage list & # esxcli vsan storage automode set --enabled false
3.To remove the specific SSD disk used in VSAN configuration, run this command:
# esxcli vsan storage remove -s [SSD-DEVICE-ID] & # esxcli vsan cluster get
4. To disconnect the host from the VSAN cluster, run this command:
# esxcli vsan cluster leave & # esxcli vsan cluster get
From the Hosts tab of the cluster, ensure no ESXi host is part of the VSAN cluster. After confirmation, right-click on the cluster and then click Remove
Removing a disk group in a 3 node cluster with only 3 disk groups is not possible.
vSan requires at least 3 disk groups spread over at least 3 hosts.
You could try to add a new disk group before removing the old one.
I have a similar problem, however the error message is this:
A general system error occurred: Disk is not mounted on this host
Because I took away the disks before make the scale down from vCenter, therefore I can't remove the disk group.
Someone can tell me how can I fix this issue?
Thanks!
Hello,
Your issue sounds different from original post here.
Anyway, what are you trying to do here, remove disks/disk-groups?
When you SSH to the host, what is the output from:
# vdq -q
# localcli vsan storage list
"Because I took away the disks before make the scale down from vCenter"
Not sure what you mean by this.
Bob
WE can remove the disk if the disk is in absent state
login in to the respective host
type vdq -q
Get the disk uuid which shows as absent
type the cmd below
esxcli vsan storage remove -u "UUID of the disk"
Now disk group will be removed
From Gui add new disk group and select the ssd and hdd and create the disk group