kiwistag
asked on
Datastore disappeared after Logical Disk resize on ESXi 5
As mentioned in Question Q_28363425, I migrated a RAID Array from RAIS 1+0 to RAID 5. After it's completion I extended the Logical Disk from roughly 1.8TB to 2.8TB (4x 1TB disks approx).
Now the datastore is gone and not viewable in ESXi.
Looking back I now realise that what has happened is that as the original Logical Disk was in MBR, that now it's over the 2TB limit it's unreadable... Is there any way to revert it or change it without data lost??
Now the datastore is gone and not viewable in ESXi.
Looking back I now realise that what has happened is that as the original Logical Disk was in MBR, that now it's over the 2TB limit it's unreadable... Is there any way to revert it or change it without data lost??
ASKER
.. For a GPT disk yes, but as the disk/partitions are still in MBR I'm in trouble.
This system is an upgrade from ESXi 4.1 so doesn't have a GPT format (let alone UEFI).
This system is an upgrade from ESXi 4.1 so doesn't have a GPT format (let alone UEFI).
ASKER
gPartEd sees all partitions which is a start. I'm looking at here for assistance..
https://communities.vmware .com/threa d/412601
https://communities.vmware
Okay, this was VMware vSphere 4.0 partition before. You question title suggests ESXi 5.0.
This is the issue.
Has the partition been expanded, as you have just expanded the logical disk?
I would shrink back the VMFS partition, to under 2TB if possible.
Did you take backups before the alteration?
This is the issue.
Has the partition been expanded, as you have just expanded the logical disk?
I would shrink back the VMFS partition, to under 2TB if possible.
Did you take backups before the alteration?
ASKER
The system has been ESXi for the last year, however beforehand was ESXi 4.1.
The logical disk has been expanded to 3TB, but the partition has not.
The VMFS partition is 1.8TB still, however ESXi 5 cannot read it in this state, I can only assume that as the (RAID LV) disk is bigger than 2TB and it is not GPT that ESXi doesn't know how to read it.
By using the MightyCare Recovery Disk, with it's later FUSE system supporting VMFS-5, I can now read the disk.
I'm copying it now to a NFS share (via SCP - since there are issues with our NAS/SAN and the Linux NFS version it seems), but at 5MBps it's going to be slow copying a main VM of 1.4TB... :/
As for backups, it looks like 1x didn't take and the other one (split-site) is 4-5 days old..
The other VM on there hasn't backed up successfully this week also. At least it's not as mission critical..
The logical disk has been expanded to 3TB, but the partition has not.
The VMFS partition is 1.8TB still, however ESXi 5 cannot read it in this state, I can only assume that as the (RAID LV) disk is bigger than 2TB and it is not GPT that ESXi doesn't know how to read it.
By using the MightyCare Recovery Disk, with it's later FUSE system supporting VMFS-5, I can now read the disk.
I'm copying it now to a NFS share (via SCP - since there are issues with our NAS/SAN and the Linux NFS version it seems), but at 5MBps it's going to be slow copying a main VM of 1.4TB... :/
As for backups, it looks like 1x didn't take and the other one (split-site) is 4-5 days old..
The other VM on there hasn't backed up successfully this week also. At least it's not as mission critical..
I think your solution is this is the best course of action, rather than meddle with the logical disk!
SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
ASKER
Solution is the prevention in a way, ensure backups are consistent.
The recovery was a rather extended process in this scenario.
The recovery was a rather extended process in this scenario.
see here
https://www.vmware.com/pdf/vsphere5/r50/vsphere-50-configuration-maximums.pdf