GPT PMBR size mismatch on data drive

Hello all,

A short while ago, my system disk failed and I fixed it in the end by cloning it using dd.

This evening, I saw the same symptoms (disk write-protected) so I cloned the disk again. There were still problems booting so I ran fdisk and agreed to all the fixes. Now when I boot up, I get an error, and on running fsck, I get this - please see the entry in red. This is one of my data disks.

Can anyone help me with this, please?

Thanks,

Nick.

EDIT: Also, forgot to say that when I try to log into the web GUI, I get a "502 - Bad Gateway" banner (in red) at the bottom of the page.

root@omvper210up:~# fdisk -l
Disk /dev/sdb: 558.91 GiB, 600127266816 bytes, 1172123568 sectors
Disk model: ST3600057SS
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disklabel type: gpt
Disk identifier: 63973638-7AB1-443D-AB0D-3EEEB94DB675

Device     Start        End    Sectors   Size Type
/dev/sdb1   2048 1172123534 1172121487 558.9G Linux filesystem


Disk /dev/sda: 4.55 TiB, 5000981078016 bytes, 9767541168 sectors
Disk model: ST5000LM000-2AN1
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 4096 bytes
I/O size (minimum/optimal): 4096 bytes / 4096 bytes
Disklabel type: gpt
Disk identifier: 333B7F94-8FB1-43B5-9398-EFA310DF8122

Device     Start        End    Sectors  Size Type
/dev/sda1   2048 9767541134 9767539087  4.5T Linux filesystem
GPT PMBR size mismatch (30629375 != 31260671) will be corrected by write.
The backup GPT table is not on the end of the device.


Disk /dev/sdc: 14.91 GiB, 16005464064 bytes, 31260672 sectors
Disk model: Cruzer Blade
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disklabel type: gpt
Disk identifier: D36D99AF-FC10-49D3-8BFA-A0F958C7F854

Device        Start      End  Sectors  Size Type
/dev/sdc1      2048  1050623  1048576  512M EFI System
/dev/sdc2   1050624 28628991 27578368 13.2G Linux filesystem
/dev/sdc3  28628992 30627839  1998848  976M Linux swap

Update: For some reason, it has fixed itself overnight. Odd.

Zitat von crashtest

Did it occur to you that you might have cloned some corrupt files?

Did it (fix itself)? Going forward, I wouldn't be surprised if odd things happen from time to time.

_________________________________________________________

If it was me, I'd rebuild. With a known good install; I'd clone my boot drive when it's working properly, before there's a boot drive failure. You'd have known good backup.

Yes, I think this may have happened (corrupt files), but on the second clone only, as I had some errors on first boot (the ones that I ran fdisk on. The first time I cloned it, this didn't happen and I still have that disk, so i will clone it again. I think the first failed disk was just a write-protect problem, but the second had other issues, it would seem.

The odd things you mentioned seem to be happening as I'm getting regular emails reporting this:

"/etc/cron.daily/openmediavault-pending_config_changes:

Segmentation fault"

I'll try cloning what I believe to be the good disk and report back. I'm only doing this as I don't know how to rebuild things as you suggested.

Thanks for your help,

Nick.

Another update.

Well, the clone seems good this time. I got a list of entries regarding orphaned files of which I vowed to remember the exact message but, as usual, a short time later I have forgotten. My bad.

After those, it booted straight into OMV. I now also have access to the web GUI again without the 502 message and I am performing an update now. I'll see how it goes for a week and if all is well I will clone the system disk to avoid this happening again. I'll resolve this thread then, too.

Just for other people's information, I think that the past two disk issues (with write-protect errors) are due to me buying cheap (SanDisk branded) USB drives off of the 'Bay. I'm not doing this again. Also, I originally was not using the flash memory plugin.

Thanks, crashtest for your help,

Nick.

Zitat von crashtest

When flash media wears out, it goes "read only". This give the appearance to the OS of "write-protected". Until the flash devices controller detects the problem and sets the entire device to "read only", some individual storage locations (usually 4k in size) refuse to write. This means that file updates did not write or may have been "partially" written creating the potential for system files to be corrupted and / or out of sync.

What may have been written, may not have been fully written, etc., can only be a matter of speculation but if it's kernel or system file related, I wouldn't trust EXT4's journaling feature to completely clean it up.

Your call.

OK, thank you for explaining this.

How would I go about rebuilding like you said earlier?

Thanks,

Nick.

Ah, yes - I know how to install it to USB but what changes do I need to make so that it hooks up to my data drives and everything works?

I have data, Piwigo and Nextcloud on my data drives, together with all my docker stuff.

Sorry if this is vague but happy to detail anything else you may need.

Thanks,

Nick.

OK, many thanks and that all seems to make sense to me.

I'll try a rebuild as soon as I get the time and I'll post back here if when I get stuck, or (if by some other-worldly miracle); I manage to do it all by myself. I'll probably continue to dd my clones but not before checking out the link you sent me - that could be easier!

Thank you again,

Nick.

Hi,

Thank you for this - I haven't got around to doing anything yet but all my Dockers are already on the data drives.

All the best,

Nick.