6.7 vs 6.5U2 - VMFS Version Differences?

[richard612]

I've got a weird one.  I have a 6.7 system with six local datastores, all attached to the same HP P420 controller.  All datastores were created under the RTM/gold build of 6.7.  None of them have been resized or modified.  ESXi was installed onto a USB key and I've been booting from the key ever since.

I then switch keys and boot the system into 6.5U2.  Only four of the six datastores are recognized.  The remaining two disks appear completely empty.  The 6.5U2 installer exhibits the same behavior -- it sees the disks and it thinks they're empty.  What gives?

I'm trying to pull off a "downgrade" by simply switching to the 6.5U2 key then re-configuring everything from scratch.  I'm hoping to preserve the datastores -- and their data!

[a_p_]

I assume that you've configured the P420 to presented 6 logical volumes to ESXi.

Please run

partedUtil getptbl /vmfs/devices/disks/....

for each of the 6 logical volumes, and post the commands including their output.

In addition to this, run esxcli storage vmfs snapshot list to see whether this command returns something.

André

[richard612]

Bingo.  I get this on two of the six volumes.

Error: Function not implemented during read on /dev/disks/naa.600508b1001c71f79961c1be1afe0576

Error: The primary GPT table states that the backup GPT is located beyond the end of disk. This may happen if the disk has shrunk or partition table is corrupted. Fix, by writing backup table at the end? This will also fix the last usable sector appropriately as per the new reduced size. diskPath (/dev/disks/naa.600508b1001c71f79961c1be1afe0576) diskSize (2344160432) AlternateLBA (2344225967) LastUsableLBA (2344225934)

Warning: The available space to /dev/disks/naa.600508b1001c71f79961c1be1afe0576 appears to have shrunk. This may happen if the disk size has reduced. The space has been reduced by (65536 blocks). You can fix the GPT to correct the available space or continue with the current settings ? This will also move the backup table at the end if it is not at the end already. diskSize (2344160432) AlternateLBA (2344225967) LastUsableLBA (2344225934) NewLastUsableLBA (2344160398)

Error: Can't have a partition outside the disk!

Unable to read partition table for device /vmfs/devices/disks/naa.600508b1001c71f79961c1be1afe0576

Any recommendations?  Is partedUtil fixGpt safe?

[EsxTest]

I have exactly the same problem.
My server: HP DL380p Gen8 Server - HPE Smart Array P420i - RAID-5 volume

I noticed this problem when upgrading from ESXi 6.0.2 to 7.0.3 but I think the problem was already there.

Suggested solution:

partedUtil fixGpt /vmfs/devices/disks/naa.600508b1001c2ae149275f04ed006297 

but I get the error:

FixGpt tries to fix any problems detected in GPT table. Please ensure that you don't run this on any RDM (Raw Device Mapping) disk. 
Are you sure you want to continue (Y/N): Y 

Error: Function not implemented during read on /dev/disks/naa.600508b1001c2ae149275f04ed006297
Retry/Ignore/Cancel?

Have you found a solution?
Thank you in advance.

[EsxTest]

I found the solution (working for me!) and I recevered old datastore.
Probably the problem was because new ESXi install is in another harddrive and previous install was in the same drive of the datastore
So, I list the datastore with command:

esxcli storage vmfs snapshot list

[root@localhost:~] esxcli storage vmfs snapshot list
512ffa9c-b397085c-5a0f-d89d6729b62c
Volume Name: SAS_15RPM
VMFS UUID: 512ffa9c-b397085c-5a0f-d89d6729b62c
Can mount: true
Reason for un-mountability:
Can resignature: true
Reason for non-resignaturability:
Unresolved Extent Count: 1

after, I get the VMFS UUID and:

esxcfg-volume -M 512ffa9c-b397085c-5a0f-d89d6729b62c

Persistently mounting volume 512ffa9c-b397085c-5a0f-d89d6729b62c

And old datastore is added and works perfectly.