I've messed up my array, is there any way back?

[Suse User]

It's serves me right for trying to install a new parity disk whilst working on the phone.

 

I wasn't paying attention and didn't notice that my Disk1 had a red dot by it after I stopped the array.  I clicked Format new disk(s) thinking I would just be preparing my new parity disk.

 

I stopped the formatting and now I have "too many wrong/missing disks"

 

I have my original parity disk that I'd just swapped out plus all the other disks from the array so I *should* be able to rebuild disk1, but the interface tells me "Wrong Disk" for the Parity volume.

 

Can I manually edit a file on the flash to tell Unraid to use my Previous parity volume?

 

 

The disk 1 contains all the licences for all the software I've ever bought, together with out Photos going back 20 years.

 

Thanks,

 

Mark.

 

PS Using 5.0-RC11

[bkastner]

First, back up your USB drive to be safe.

 

The, you should be able to run New Config from the GUI, then reassign the old parity drive and all your data drive and mark that Parity is Valid so that it doesn't try and rebuild it. Once the array come up, then you can go and put in a new drive for disk1 to have it rebuild.

 

I am 99% sure this is correct, but I'd wait for confirmation just to be safe.

[Suse User]

First, back up your USB drive to be safe.

 

The, you should be able to run New Config from the GUI, then reassign the old parity drive and all your data drive and mark that Parity is Valid so that it doesn't try and rebuild it. Once the array come up, then you can go and put in a new drive for disk1 to have it rebuild.

 

I am 99% sure this is correct, but I'd wait for confirmation just to be safe.

 

Thanks for your reply, as you suggest I think I'll wait for confirmation to be sure...

 

I did find a bit in the Wiki about swap-disable

 

If the replacement disk is larger than your parity disk, then the system permits a special configuration change called swap-disable.

For swap-disable, you use your existing parity disk to replace the failed disk, and you install your new big disk as the parity disk:

Stop the array.

Power down the unit.

Replace the parity hard disk with a new bigger one.

Replace the failed hard disk with you old parity disk.

Power up the unit.

Start the array.

When you start the array, the system will first copy the parity information to the new parity disk, and then reconstruct the contents of the failed disk.

 

But as my previous Parity disk is not recognised as valid in the array I don't think that will work for me.

 

Thanks,

 

Mark.

[dgaschk]

See here: http://lime-technology.com/forum/index.php?topic=32491.msg298738#msg298738

[Suse User]

That looks like it's going to be very helpful, Thanks...

 

From that thread:-

See here: http://lime-technology.com/forum/index.php?topic=32491.msg298738#msg298738

 

Just so I'm Crystal Clear..

 

> 1. Execute Utilities/New Config.

>

> 2. on Main, assign all your drives, being very careful to assign Parity and your new disk (disk10) correctly.

>

> 3. From a console or a telnet session type this command:

>

> mdcmd  set  invalidslot  10

>

> (the 10 corresponds to disk10)

>

> 4. Click 'Start' on the webGui.

>

> What should happen now is array gets started with disk10 reconstruct in process.

>

> IMPORTANT: between steps 3 and 4 do NOT refresh your browser or navigate to any other pages in the webGui - just click the Start button which is already being displayed there (if you navigate to a different page or even refresh the browser after step 3 it will "cancel" the effect of that 'mdcmd' and result in your parity disk getting written - not good).

 

In my case I'd set:-

 

mdcmd  set  invalidslot  1

 

As my "Disk1" is the one that needs rebuilding from the rest?

[dgaschk]

Yes.

[garycase]

That should work ... and since you apparently don't have any backups hopefully it will.    It's really surprising to me you wouldn't have backups of something as important as "... all the licences for all the software I've ever bought, together with our Photos going back 20 years."

 

Hopefully that won't be an issue going forward if you recover all the data 

[Suse User]

That should work ... and since you apparently don't have any backups hopefully it will.    It's really surprising to me you wouldn't have backups of something as important as "... all the licences for all the software I've ever bought, together with our Photos going back 20 years."

 

Hopefully that won't be an issue going forward if you recover all the data 

 

Hi there,

 

There's no excuse for it.  I'm amazed I don't have a backup of the photos, I thought I did on an external USB HDD but looking at that now it appears it got used for emergency repairing a family members computer a year or so ago.  I don't remember using that disk for backup, but that's what's on it.

 

All the current photos (past few years) are also on main computer too, but I felt safe thinking I had 2 backups of the old ones.

 

It appears I have problems with my UPS, I'm getting intermittent power fluctuations.

 

I followed those 'mdcmd' instructions and it allowed me to configure my array again correctly, but on restarting another disk that was previously fine (Disk18 with green dot) appeared to be trying to do a data recovery but failing with contest errors and disk1 wasn't being written to at all.  I stopped the array again and Disk18 was not present.

 

I'm getting in a right mess here.

 

I'm going to buy another 3TB disk (same size as my old parity) to swap out for Disk1 & use the backup of my USB key (I took before starting this) to try the mdcmd approach again.

 

I did manage to do a reiserfsck (readonly) on my disks after a reboot and disk2 and disk18 won't run that test because of "Bad Block 2"  Disk18 is (currently) non-critical backups of other computers, I'm not overly concerned about that.  I do recall there is a method to recover data on a disk that has a corrupted file system.  I'm not sure if that is the reiserfsck --rebuild-tree, even if --rebuild-sb may work.  Maybe it's a different command entirely....  but I'm beginning to think this will be a better chance of getting my pics back.

 

Once my new Disk arrives I'll attempt the 'mdcmd' approach again as well.

 

Thanks for your help

 

Mark.

[garycase]

Good luck, although with two failed disks (1 & 18) you simply won't be able to do a rebuild, so your only real hope is that disk1 can be read on another system.

 

I'd take out disk1, attach it to a PC's native SATA port, and see if you can read it using the free LinuxReader.  [ http://www.diskinternals.com/linux-reader/ ]    If so, copy whatever data you can recover to another disk on your PC.

 

If that doesn't work; put disk1 back in your UnRAID PC and try doing a reiserfsck run on it.

 

If neither of those work, I'd send it off for professional data recovery.  Gillware is very good at this;  has a no-recovery, no-fee policy; and is reasonably priced (by data recovery standards).  [ https://gillware.com/ ]

 

[Suse User]

Good luck, although with two failed disks (1 & 18) you simply won't be able to do a rebuild, so your only real hope is that disk1 can be read on another system.

 

I'd take out disk1, attach it to a PC's native SATA port, and see if you can read it using the free LinuxReader.  [ http://www.diskinternals.com/linux-reader/ ]    If so, copy whatever data you can recover to another disk on your PC.

 

If that doesn't work; put disk1 back in your UnRAID PC and try doing a reiserfsck run on it.

 

If neither of those work, I'd send it off for professional data recovery.  Gillware is very good at this;  has a no-recovery, no-fee policy; and is reasonably priced (by data recovery standards).  [ https://gillware.com/ ]

 

Thank again for all your help.  It turns out that the external backup I thought I had of my Up to 2004 photos still exist, just on a different disk to the one in the box marked "Photos" .  I said I had my recent (past few years) photos still on my computer, it turns out I have everything since 2004 for there.  I'd have guessed I did the archive/clearout 2 or 3 years ago.  Turns out it was 10 years ago!  No photos lost.  I have the actual licence numbers in a database program, but the licence files required by the odd app are still currently lost.

 

I will still try the recovery methods you have suggested

 

Thank,

 

Mark.

[garycase]

Glad you found the backups (or at least the key photos).

 

It always surprises me how many folks consider their UnRAID array as not requiring backups.  NO RAID system is a substitute for backups.    I have 3 UnRAID arrays -- one for media; one for everything else; and the 3rd which backups up all of the first two arrays (over 40TB).    Yes, it costs a bit; but my view is if the data's important enough to build a fault-tolerant server to store it on; it's also important enough to back up.

 

[jonp]

Glad you found the backups (or at least the key photos).

 

It always surprises me how many folks consider their UnRAID array as not requiring backups.  NO RAID system is a substitute for backups.    I have 3 UnRAID arrays -- one for media; one for everything else; and the 3rd which backups up all of the first two arrays (over 40TB).    Yes, it costs a bit; but my view is if the data's important enough to build a fault-tolerant server to store it on; it's also important enough to back up.

And its never been cheaper or easier to backup your data either...

[Suse User]

I've actually recovered all critical data to the cache disk by manually mounting the device volumes.

 

 

I used:-

 

mkdir /mnt/cache

mount /dev/sdb1 /mnt/cache

 

mkdir /mnt/disk1

mount /dev/sdd1 /mnt/disk1

 

cd /mnt/disk1

cp myfolder /mnt/cache -R

 

And it appears to have copied correctly.

 

In order to recover more (not critical) data from the two damaged disks I'd need more space than I have available.  Is there a command line I can use to copy entire folders to another UnRaid server on my network?

 

Thanks,

 

Mark.