36 minutes ago, trurl said:

According to your previous diagnostics, sdh was not the 6TB disk formerly assigned as disk7. It was a 2TB unassigned disk serial ending 5257

yes i tried to copy the 1.5tb content on the 6tb disk to a 2 tb disk

14 minutes ago, JorgeB said:

Sorry, this thread is getting confusing. Assuming that was the correct disk, if ddrescue cannot copy anything, try copying what you can from the emulated disk.

i'm like 3 pages asking what to do lol

i have 2 disks with problems

1. disk 2 - 2TB, is in the array, extended smart test ok, cant mount in the array. running check file system for like 3 time.. still running fix there

2. disk 7 - 6 TB is not in array, is at unassigned devices, cant mount there, I do see (using explorer) 1.5 TB data on it, tried to copy/ddresuce without luck. Extended smart test ok

Spare/new disks available:

• 2 TB which is now at unassigned devices/mounted

• 10 TB to upgrade parity later (which is 8 tb now)

• 10 TB new lying here

So i want to rescue as much as possible.

What are the steps after i finished the fix script running on disk 2 and hopefully i can get somewhere today...

46 minutes ago, skank said:

disk 7 - 6 TB is not in array, is at unassigned devices, cant mount there, I do see (using explorer) 1.5 TB data on it, tried to copy/ddresuce without luck. Extended smart test ok

Since the disk itself appears to be completely dead, copy what you can from the emulated disk.

46 minutes ago, skank said:

disk 2 - 2TB, is in the array, extended smart test ok, cant mount in the array. running check file system for like 3 time.. still running fix there

For this one, if xfs_repair cannot fix the filesystem, you can try a file recovery app like UFS Explorer; the free trial should show what it can recover.

30 minutes ago, JorgeB said:

Since the disk itself appears to be completely dead, copy what you can from the emulated disk.

With what command?

3 hours ago, skank said:

disk 7 - 6 TB is not in array, is at unassigned devices, cant mount there, I do see (using explorer) 1.5 TB data on it

If it isn't mounted there is no way to access its contents. What you are seeing from Windows is emulated disk7.

And since you are still checking disk2, there is no way to even get that since the array isn't started in normal mode.

You could skip disk2 for now, and just rebuild disk7, assuming emulated disk7 is indeed mountable. New Diagnostics with the array started in normal mode would answer that question.

The last, or one of the last, diags posted do show emulated disk7 mounting:

Jan 14 09:29:09 Unraid emhttpd: shcmd (82): mkdir -m 0666 -p /mnt/disk7
Jan 14 09:29:09 Unraid emhttpd: shcmd (83): mount -t xfs -o noatime,nodiscard,nouuid /dev/md7p1 /mnt/disk7
Jan 14 09:29:09 Unraid kernel: XFS (md7p1): Mounting V5 Filesystem 6b0c6369-f0e2-4864-8cce-b9d6548007ea
Jan 14 09:29:09 Unraid kernel: XFS (md7p1): Starting recovery (logdev: internal)
Jan 14 09:29:09 Unraid kernel: XFS (md7p1): Ending recovery (logdev: internal)
Jan 14 09:29:09 Unraid emhttpd: shcmd (84): xfs_growfs /mnt/disk7
Jan 14 09:29:09 Unraid kernel: xfs filesystem being mounted at /mnt/disk7 supports timestamps until 2038-01-19 (0x7fffffff)
Jan 14 09:29:09 Unraid root: meta-data=/dev/md7p1             isize=512    agcount=6, agsize=268435455 blks
Jan 14 09:29:09 Unraid root:          =                       sectsz=512   attr=2, projid32bit=1
Jan 14 09:29:09 Unraid root:          =                       crc=1        finobt=1, sparse=1, rmapbt=0
Jan 14 09:29:09 Unraid root:          =                       reflink=1    bigtime=0 inobtcount=0 nrext64=0
Jan 14 09:29:09 Unraid root:          =                       exchange=0   metadir=0
Jan 14 09:29:09 Unraid root: data     =                       bsize=4096   blocks=1465130633, imaxpct=5
Jan 14 09:29:09 Unraid root:          =                       sunit=0      swidth=0 blks
Jan 14 09:29:09 Unraid root: naming   =version 2              bsize=4096   ascii-ci=0, ftype=1, parent=0
Jan 14 09:29:09 Unraid root: log      =internal log           bsize=4096   blocks=521728, version=2
Jan 14 09:29:09 Unraid root:          =                       sectsz=512   sunit=0 blks, lazy-count=1
Jan 14 09:29:09 Unraid root: realtime =none                   extsz=4096   blocks=0, rtextents=0
Jan 14 09:29:09 Unraid root:          =                       rgcount=0    rgsize=0 extents
Jan 14 09:29:09 Unraid root:          =                       zoned=0      start=0 reserved=0

Assuming it still does, you can copy the data locally to another array or unassigned disk, or even over the LAN.

Yesterday evening i ran the check file system on disk 2 then after it i saw a button to fix it. So i pressed fix it. It was still running.. and i went to sleep. this morning it wasnt running anymore. But the button "check file system" was there again... i dont get it. It still isnt mountable.

Turned off server then

Now i turned on.. everything is still the same.

I just started copying over the data from the emulated disk 7 (1.5TB ), its not mounted... but it is copying and i do see the unassigned but mounted 2tb disk filling up...

LEts hope it finishes copying then i think i got all data from disk 7 ?

If the emulated disk7 is mounting, it should have all the data that was on that disk.

1 hour ago, skank said:

started copying over the data from the emulated disk 7 (1.5TB ), its not mounted

What you actually mean it it's not assigned. It must be mounted or you couldn't see its files. "Mount" is when the OS loads the filesystem to make its contents accessible.

An emulated disk can be mounted even if the physical disk has been destroyed.

4 minutes ago, trurl said:

An emulated disk can be mounted even if the physical disk has been destroyed.

Assuming the emulated filesystem is mountable.

1 hour ago, trurl said:

What you actually mean it it's not assigned. It must be mounted or you couldn't see its files. "Mount" is when the OS loads the filesystem to make its contents accessible.

An emulated disk can be mounted even if the physical disk has been destroyed.

What i meant there is, that the disk in the array page, isnt mounted, cause it was not assigned

I dont know how to mount/unmount emulated disk. Anyway its still copying

Ok so i copied over data fom disk 7 to the spare 2tb

Now what?

Try again to repair disk 2

Or do i take out disk 7, format it under windows -> take out disc 2 from array, and put back disk 7 instead?

43 minutes ago, skank said:

Try again to repair disk 2

You can try the result will be likely the same.

You can also try running UFS Explorer on it; the free trial should show if it can recover the data.

2 minutes ago, JorgeB said:

You can try the result will be likely the same.

You can also try running UFS Explorer on it; the free trial should show if it can recover the data.

Downloading right now

Got it running, lets see

Edited January 15Jan 15 by skank

3 hours ago, skank said:

dont know how to mount/unmount emulated disk

Unraid mounts the emulated disk when you start the array (unless it is unmountable).

Ok tonight i will see if i can recover data from disk 2 with UFS explorer by connecting the physical disk to my laptop.

While i do that, and since data from disk 7 is on my spare disk, i assume i can move further on with my unraid machine?

Can i do the following safely?

Make a new config -> disk 2 and disk 7 out...

So array comes from:

disk 1,2,3,4,5,6,7 to new array disk 1,2 (old 3), 3 (old 4),4 (old 5), 5 (old 6) ? Without losing the data on those disks?

Can i immediately swap my parity for a 10tb disk?

Or do i need to build parity first on the new config and then swap?

Or will i lose my data on the old disks if i do new config?

38 minutes ago, skank said:

disk 1,2,3,4,5,6,7 to new array disk 1,2 (old 3), 3 (old 4),4 (old 5), 5 (old 6) ? Without losing the data on those disks?

Yes, that should be fine; it should reimport all the existing data disks. You can assign the new parity and sync with that one.

12 minutes ago, JorgeB said:

Yes, that should be fine; it should reimport all the existing data disks. You can assign the new parity and sync with that one.

Hi ok cool will do that.

If UFS explorer on disk 2 fails, i will lose data anyway i think.

So i can safely do the new config with a new 10tb parity

do i need to scan like this on disc 2 with ufs

Don't forget to check XFS

5 hours ago, JorgeB said:

Don't forget to check XFS

Yes i am (changed after screenshot), its running now

Stiil having problems

I did a new config with parity 10tb

i added a new disk 6 with 10tb free for data (HGST drive)

All my data disks are xfs filesystem

I started building parity and then i saw the last disk 6 10TB was filesystem ntfs

I let him finish parity sync first,

i then stopped aray, took out disk 6 and formatted to xfs using command line

I readded it to array

and now data rebuild is being done, but i see its back to ntfs

What am i doing wrong? Disk settings are set to default xfs..

i couldnt add another new disk to slot 7 either.. so i will have to resync again after the rebuild.. crap its taking ages

Edited January 17Jan 17 by skank

And another problem, i see recoverd things... with UFS explorer

However when i want to recover the things...

i get this

so much for the trial.; I think i downloaded the professional version, im not sure if i download the standard version if i will get the same reults and trial is possible too?

Edited January 17Jan 17 by skank

Probably the NTFS drive has multiple partitions.

3 hours ago, skank said:

formatted to xfs using command line

You should use the webUI to erase the drive, and then use the webUI to format it to XFS.

8 hours ago, trurl said:

Probably the NTFS drive has multiple partitions.

You should use the webUI to erase the drive, and then use the webUI to format it to XFS.

So i should stop rebuilding now?

Yeh probably multiple partitions,

Can you tell me exact steps please?

Cause i didnt see an erase and format button.

Its still rebuilding

What are the exact steps please to get it to xfs in the array , and add another 8 tb drive (already in xfs)