Jump to content
XPEnology Community

MAJOR failure - XPE reports as 16383 Petabyte volume etc


illisis

Recommended Posts

Hey guys

 

I have XPE running on Hyper-V and use an ISCSI target from that.. XPE had 5 disks using passthrough to XPE then one failed. I replaced that and I believe there was a power outage or something caused the volume to go haywire.. Now the volume reports as 16+ PB and cpu hits 100% because storage manager hangs/crashes

 

basically I need a miracle.. =/

 

screenshots

8b3Kc0N.jpg

 

X52krGR.jpg

Link to comment
Share on other sites

5 disk, I'll assume that was a RAID5 volume?

 

is the OLD drive that you replaced still "sort of" work?

"sort of" work means it still spins up, and can read partial data, it'll probably have a lot of bad sectors, but at least it can still read something from it.

 

if the old drive is dying, but not completely dead.

the best way is to use Clonezilla and clone the dying drive into an image or directly to another HDD then take that HDD into the RAID5

 

then you might get a better chance at recovering the RAID if you can plug back in the "original" drive that was replaced

Link to comment
Share on other sites

For the future planning:

 

Set up set up 2 servers to mirror each other if you value your data, specially if you use RAID5.

 

Best practice, never got with RAID5, use RAID6 as the bare minimum.

RAID10 is the safest, fastest, and relatively "low cost" compared to RAID50 or RAID60 setups.

 

Even for my hobby home setup, I use a minimum of RAID10 on each mirror server.

so if a few HDD dies, it's not a big deal

if the whole server dies, it's not a big deal

but if both servers decided to die at the same time... well I'm royally screwed.

 

I've had 3 HDD die on me from the same RAID volume, I'm glad it was RAID10

they just decided to Expire at the same date, they were purchased in a batch of 12, the remaining 11 drives continues to work fine today many years after the warranty has gone by already.

 

it's a luck of the draw, once you are out of the warranty period, expect HDD to die randomly.

 

I always keep a few on had for spare to hot swap for when a disk dies.

 

That's why that day when 3 died suddenly it caught me off guard, it it was a RAD5 or RAID6 i'd have lost the data.

 

Actually since that day I added a 2nd server to mirror the 1st server, both running on RAID 10

 

You never know when a weird struck of luck will hit you with a multi drive death that can even paralyze a RAID10 :razz:

Link to comment
Share on other sites

5 disk, I'll assume that was a RAID5 volume?

 

is the OLD drive that you replaced still "sort of" work?

"sort of" work means it still spins up, and can read partial data, it'll probably have a lot of bad sectors, but at least it can still read something from it.

 

if the old drive is dying, but not completely dead.

the best way is to use Clonezilla and clone the dying drive into an image or directly to another HDD then take that HDD into the RAID5

 

then you might get a better chance at recovering the RAID if you can plug back in the "original" drive that was replaced

 

Actually, It was set on SHR. I believe thats the equivalent?

 

Story was that I had storage spaces setup using these disks and I had a 1.5tb drive die(Seagate) I replaced that drive with a 4tb and it wasn't rebuilding for whatever reason so I copied off all the data and recreated the storage space when I was putting all the data back on the new space a 1tb drive died(Seagate) it was also thrashing the disks so I decided to move to a virtual XPE with passthrough disks and just iscsi to it. Then another 1tb drive died(seagate) and thats the one that I replaced with yet another 4tb then all this happened

 

The old drive has just been sitting there the reason it was dying was the heads were acting up so other than it carrying on and ticking every now and then it should be usable so I tried to plug it back in last night and booted with no resolution to the problem.

Link to comment
Share on other sites

If I read you right,

it sounds like the drive you took out originally still sort of work, but the other 2 drives you had in there just crashed one after the other, while the new drive that replaced the original drive you took out was still syncing up (Repairing) to the original RAID... which unfortunately failed to complete because the other drives died one after the other, leaving you with a broken RAID.

 

I'm not sure you can recover from that, perhaps some of the other guys that frequent this forum might have other ideas that can help, but personally I'm out of ideas.

Link to comment
Share on other sites

Ok i've plugged in the faulty drive with no luck..

 

Does anyone know how do un-mount /volume1?

 

Here is a screenshot of all my disks:

From what I can determine is that the newest 4TB drive hasn't been touched.. And the array doesn't look complete on the oldest 4tb drive

There is two 1TB Seagate drives connected, one wasn't used whet the array was setup - don't know why it has those partitions

The other 1TB drive is the dead one from this array

There are 4 disks with extended partitions which is 4/5 disks which should give me something back?!

 

9GIIXJA.jpg

Link to comment
Share on other sites

You're probably better off trying to mount the array in standard linux. I lost my array once a while back, and from a lot of reading how linux raid works, and help from someone on the TestDisk forums I was able to recreate my array w/o having any partition tables, backup my data and start fresh. It took me like 6 months until I got my data back - now I make backups :wink: From my experience once Synology marks your array as failed, ect, it's very hard to get it to work in DSM again (unless you work for synology - they don't tell you what they do when they SSH into your box to recover arrays).

 

Anyway, I outlined everything I did back then, and it can be found here... it may or may not be of any help: https://forum.cgsecurity.org/phpBB3/fou ... t2600.html . All I can say is take your time and do a lot of reading. Image the disks to another set if you have spares, and work with the imaged drives in case you make it worse. If you have backups, it's probably easier to format and start over... unless your just curious and looking for a huge challenge.

Link to comment
Share on other sites

You're probably better off trying to mount the array in standard linux. I lost my array once a while back, and from a lot of reading how linux raid works, and help from someone on the TestDisk forums I was able to recreate my array w/o having any partition tables, backup my data and start fresh. It took me like 6 months until I got my data back - now I make backups :wink: From my experience once Synology marks your array as failed, ect, it's very hard to get it to work in DSM again (unless you work for synology - they don't tell you what they do when they SSH into your box to recover arrays).

 

Anyway, I outlined everything I did back then, and it can be found here... it may or may not be of any help: https://forum.cgsecurity.org/phpBB3/fou ... t2600.html . All I can say is take your time and do a lot of reading. Image the disks to another set if you have spares, and work with the imaged drives in case you make it worse. If you have backups, it's probably easier to format and start over... unless your just curious and looking for a huge challenge.

 

Interesting! Reading your efforts made me realise that I am further along than where you started.. I can already mount it in Parted Magic and see the directories that xpe has used.

My only problem is that I used ISCSI which I believe is the EP_DAT_0000.DAT file in the @EP directory which is 3.3TB.

 

My data is Bitlockered =/ so how can I copy and use of this in another XPE VM??

Link to comment
Share on other sites

Bitlocker is the Microsoft encryption to hard drive, just like PGP from Symantec, or similar encryption from FirstPoint.

 

Situations where the Bitlocker is used, You'll first want to De-Encrypt the data before making any repairs, you can't make repairs while data are encrypted.

Link to comment
Share on other sites

I get that I would need to decrpyt the data to get anything.

 

The Data is in the ISCSI volume which is on Volume1 on the XPE So I need to move the ISCSI database(which I believe to be EP_DAT_00000 to a new server(how do I do this is my question) and then retarget my initiator to get the data out

Link to comment
Share on other sites

Try setting up another XPE VM, make an iSCSI drive the same size as your old one. Shut it down, or disable it, and swap the file you mention above.

 

I have no idea if that will work, but it might be worth a try if you haven't done something like it yet. I use iSCSI as well, but haven't tried moving the iSCSI drives from one VM to another... But I have accidentally deleted one that I had a lot of files on.. and then much later wondered where I had those file and realized I had deleted the iscsi drive they were on :sad:

Link to comment
Share on other sites

×
×
  • Create New...