Thanks to you guys my original question has been answered - unfortunately, I now have a new problem. Apologies for the mass of text:

 

In attempting to add a newly-precleared disk to my array (my original question), one of the fixes I tried was to take the array offline and switch the new drive's hot-swap bay in case it was a cabling issue. Didn't make a difference, so I swapped everything back to the way it was. All drives were greenballed, but when I went to restart the array one disk showed as "unformatted". To my knowledge I hadn't written anything to the drive, but to be safe I followed the procedure outlined in the wiki (http://lime-technology.com/wiki/index.php?title=Troubleshooting#How_do_I_re-enable_the_drive.3F) to reconstruct that drive, since the parity drive had never been touched. All went fine with that process and the array was back up and running about 10 hours later. Fast forward to the next day, when a helpful poster advised me to reboot in order to get my new, un-added drive to show up in the management utility - that worked great, but now a *different* original data drive shows as unformatted.

 

Not sure what the next step is. Is it safe to use the reconstruct method on that drive as well, or do I risk losing data? Please forgive me if this is a stupid question, I'm pretty new to unRAID so I want to make sure I don't mistakenly do anything to make things worse. Syslog attached for your viewing pleasure, please let me know if you see anything out of order.

syslog_11_07_0928.txt

Running a reiserfsck check results in:

 

reiserfs_open: the reiserfs superblock cannot be found on /dev/md3.

Failed to open the filesystem.

 

If the partition table has not been changed, and the partition is

valid  and  it really  contains  a reiserfs  partition,  then the

superblock  is corrupted and you need to run this utility with

--rebuild-sb.

Running a reiserfsck check results in:

 

reiserfs_open: the reiserfs superblock cannot be found on /dev/md3.

Failed to open the filesystem.

 

If the partition table has not been changed, and the partition is

valid  and  it really  contains  a reiserfs  partition,  then the

superblock  is corrupted and you need to run this utility with

--rebuild-sb.

In this thread:

http://lime-technology.com/forum/index.php?topic=15385.0

you 'll find a "dd" command I gave to another user whose disks were unformatted.  We discovered the MBR on those disks had been changed so it was looking in the wrong place for the file system. 

 

You need to run the same "dd" command and post the output (or do the analysis yourself to see if the file system is where the partition start says it should be)

 

the eventual fix for them was to fix their MBR, but your issue might be different.

 

If your file system is there, but not recognized, then you'll need to run the rebuild-sb option.  If instead, the MBR points to the wrong starting sector, you;ll need to fix it.

 

Joe L.

Joe L.

Ok, this is what I get:

 

root@Tower:~# dd if=/dev/sdd count=195 | od -c -A d |  sed 30q

195+0 records in

195+0 records out

99840 bytes (100 kB) copied, 7.04578 s, 14.2 kB/s

0000000  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0

*

0000448  \0  \0 203  \0  \0  \0  @  \0  \0  \0  p 210 340 350  \0  \0

0000464  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0

*

0000496  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  U 252

0000512  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0

*

0098352  \0  \0 003  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0

0098368  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0

*

0099840

root@Tower:~#

 

I can see that this is vastly different from the results posted on the thread you linked (most prominently, "ReIsEr2Fs" doesn't even appear), but other than that I have no idea what this means. Did I mistype, or is there something more sinister going on?

Ok, this is what I get:

 

root@Tower:~# dd if=/dev/sdd count=195 | od -c -A d |  sed 30q

195+0 records in

195+0 records out

99840 bytes (100 kB) copied, 7.04578 s, 14.2 kB/s

0000000  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0

*

0000448  \0  \0 203  \0  \0  \0   @  \0  \0  \0   p 210 340 350  \0  \0

0000464  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0

*

0000496  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0   U 252

0000512  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0

*

0098352  \0  \0 003  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0

0098368  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0

*

0099840

root@Tower:~#

 

I can see that this is vastly different from the results posted on the thread you linked (most prominently, "ReIsEr2Fs" doesn't even appear), but other than that I have no idea what this means. Did I mistype, or is there something more sinister going on?

It seems to indicate there is no file system.      (in other words, it really is unformatted)

This is not a 3TB drive, is it?  (does not look like it from the MBR, but figured I'd ask anyways)

 

 

 

 

Nope, it's definitely a 2TB. I started my build with 3x 2TB drives, and the problem drive was added about a month ago when I started running out of space in order to make 4x 2TB. I've had absolutely no trouble with it up until my recent problem trying to add a fifth drive, so no idea why it would suddenly show up as unformatted. Could it be a result of my reconstructing a different drive as mentioned in my original post? At the time, though, *that* drive (my disk2) was the only drive showing as unformatted, and my current problem drive (disk3) showed as greenballed and normal up until my most recent reboot.

 

I'm completely stumped! I do have another 2TB drive precleared and ready to go (the fifth drive I was trying to add to the array when things started going wrong), so would it help at all to swap that precleared drive for my current problem disk3 and let the array reconstruct the data based on parity? Or, does this disk3 look normal (other than appearing to have eaten my files ), such that I could stop the array and restart with this disk3 in order to reconstruct my data from parity? Please let me know if you need more info, screenshots, etc. - I really appreciate all you've done to help thus far!

Nope, it's definitely a 2TB. I started my build with 3x 2TB drives, and the problem drive was added about a month ago when I started running out of space in order to make 4x 2TB. I've had absolutely no trouble with it up until my recent problem trying to add a fifth drive, so no idea why it would suddenly show up as unformatted. Could it be a result of my reconstructing a different drive as mentioned in my original post? At the time, though, *that* drive (my disk2) was the only drive showing as unformatted, and my current problem drive (disk3) showed as greenballed and normal up until my most recent reboot.

 

I'm completely stumped! I do have another 2TB drive precleared and ready to go (the fifth drive I was trying to add to the array when things started going wrong), so would it help at all to swap that precleared drive for my current problem disk3 and let the array reconstruct the data based on parity? Or, does this disk3 look normal (other than appearing to have eaten my files ), such that I could stop the array and restart with this disk3 in order to reconstruct my data from parity? Please let me know if you need more info, screenshots, etc. - I really appreciate all you've done to help thus far!

I certainly hope you are going by the model/serial number of your disks and not the three letter device name.  (Those can change from one boot to another)

 

If your array is otherwise all "green" then you can try a NOCORRECT parity check.  If it shows no errors, then you can use this "unformatted" drive to reconstuct a different drive.

Oh no, I'm going by the model/serial of the disk! I'm just using the device name/disk # to clarify which one I'm talking about - mostly in my own head, I suppose. On that topic, thanks for clearing me up about the changing drive letters in my previous thread - should've figured that was a non-issue but I appreciate the clarification nonetheless.

 

Forgive my ignorance, but is a NOCORRECT parity check the check I would run by clicking the "Check" button on the main page of the GUI, or a different kind altogether? My array is indeed showing all "green", so would I be able to reconstruct data on this problem drive by using the procedure described here? http://lime-technology.com/wiki/index.php?title=Troubleshooting#How_do_I_re-enable_the_drive.3F Or is it a better bet to use a different drive entirely?

http://lime-technology.com/wiki/index.php?title=Un-Official_UnRAID_Manual

 

In the same way, after replacing a data disk, it is recommended that a non-correcting parity check be run. This is only available on the command line after logging in via telnet or on the system console as root. To perform that test, log in and type:

 

/root/mdcmd check NOCORRECT

 

If you have unmenu installed it is also available under "Array Management"

Oh no, I'm going by the model/serial of the disk! I'm just using the device name/disk # to clarify which one I'm talking about - mostly in my own head, I suppose. On that topic, thanks for clearing me up about the changing drive letters in my previous thread - should've figured that was a non-issue but I appreciate the clarification nonetheless.

 

Forgive my ignorance, but is a NOCORRECT parity check the check I would run by clicking the "Check" button on the main page of the GUI, or a different kind altogether? My array is indeed showing all "green", so would I be able to reconstruct data on this problem drive by using the procedure described here? http://lime-technology.com/wiki/index.php?title=Troubleshooting#How_do_I_re-enable_the_drive.3F Or is it a better bet to use a different drive entirely?

No, they are slightly different.

 

The button on the GUI runs a correcting parity sync.  It updates parity.  If your data drive is failing, it it s bad thing to do, since it will overwrite parity with possibly bad data read from the misbehaving data drive.

 

The command I gave will NOT update parity (although the GUI will still look the same, and still show as if it did write to the parity disk, since the GUI does not know how to display differently for a NOCORRECT check.)

 

Joe L.

Ok! My NOCORRECT parity check finished with 0 errors. Does that mean it's safe to remove/re-add the "unformatted" disk, then reconstruct the data using parity calculation? Or should I not trust this drive and swap it for the new, precleared drive I have on hand, then preclear my problem drive for use later on?

 

I still don't get why all of a sudden my disks have been going haywire, but it's probably something I did since I had no problems whatsoever with my array until I recently attempted to add my fifth disk. Appreciate all your help, guys - I'm cautiously optimistic at this point.

Ok! My NOCORRECT parity check finished with 0 errors. Does that mean it's safe to remove/re-add the "unformatted" disk, then reconstruct the data using parity calculation? Or should I not trust this drive and swap it for the new, precleared drive I have on hand, then preclear my problem drive for use later on?

It means if you were to remove the disk and re-construct with parity in combination with the other data disks you would end up with a disk that is exactly as you have now... unformatted.  In other words, your parity reflects whatever you did to un-format the drive.  It is almost as if you cleared the wrong drive.

 

Did you , by chance, hot-plug drives into your server after booting it?  That would possibly change the device names...  (unRAID is NOT hot-plug capable)

 

I still don't get why all of a sudden my disks have been going haywire, but it's probably something I did since I had no problems whatsoever with my array until I recently attempted to add my fifth disk. Appreciate all your help, guys - I'm cautiously optimistic at this point.

It means if you were to remove the disk and re-construct with parity in combination with the other data disks you would end up with a disk that is exactly as you have now... unformatted.   In other words, your parity reflects whatever you did to un-format the drive.  It is almost as if you cleared the wrong drive.

 

Did you , by chance, hot-plug drives into your server after booting it?  That would possibly change the device names...  (unRAID is NOT hot-plug capable)

 

I don't believe so, but I honestly can't remember. Throughout this process, I've been relying strictly on the drive serial #s when making changes, so I don't think I cleared any disks other than the one I precleared in order to add it to my array (but was unable to, so it's still sitting in its bay unassigned). I've been trying to be very careful and not screw things up further in a misguided attempt to fix my problems, which is not to say it isn't possible that I panicked and did something stupid.

 

So what kind of scenario are we looking at? Are my data hosed? From everything I can see, all files/folders sitting on my array look the same as they always have, but I haven't touched anything since I started having trouble. Is there any way to get things back to normal?

So at this point I assume I'm SOL and the drive contents are toast. Is there any way to tell which files were on there so I can attempt to sort out my remaining data? It's pretty much all video files, but I have no idea what I still have and what I don't as I've been pretty much afraid to touch anything until getting this figured out. How would I go about replacing the drive - can it be formatted/assigned to the array as if it were newly precleared and ready to go, or do I need to go through the preclearing process again as if it were completely brand-new? Or is it best to swap it out for an entirely different drive?

 

I've gone through the wiki and FAQ reading anything that seems relevant, but to be honest there's a lot of information for a noob like me and I'm pretty overwhelmed. Nothing at all against LimeTech, I should have taken more time to make sure I knew what I was doing! Thanks a lot to all you forum folks for your help, I'm hopeful that I can get my array back (or at least parts of it) and continue using this great OS

From reading other forum posts with similar issues it sounds like my drive contents are gone, which is a bummer but not a tragedy since there's nothing on there that can't be replaced. Should've been more careful - lesson learned!

 

What I'm trying to figure out now is the next step: since my disk3 is showing unformatted as if it were a completely new, data-less drive, can I treat it as such (as if I'd just precleared it and am ready to assign it to the array to expand my storage capacity)? Or are there other steps involved here that differ from the new-drive process, since this drive isn't actually new to the array but only appearing as if it is?

 

Sorry for all the noob questions, but I got myself into this situation by thinking I knew what I was doing and don't want to make the same mistake twice. Basically I just want to get my array up and running again! Any ideas?