I'm currently Running windows 10.  I've had my Unraid server running for years, dealt with bad disks before, but nothing like this.  My current config consists of 6 data disks, and a parity drive.  Total available space is 16TB, 10.8 used. 

 

A week or so ago I had a drive that was showing some errors, so I replaced it with a larger drive.  Rebuild went fine, then ran a parity check to make sure all was good.  No issues there.  I didn't have any problems until I tried run the mover script to move about 40GB over from my cache drive to the data drives.  Unraid hung, the webgui became unresponsive.  I telnetted in and tried to do a clean powerdown but that didn't work.  I looked at the syslog and saw a kernel panic!  This made me think maybe the cache drive was going bad.  So I disabled the cache drive, and tried to copy directly to the share/disk.  Crashed again.  I tried again with a tiny txt file, and that worked!  So I tried copying an mp3 file of like 10MB, and that caused a crash as well! 

I read some posts stating that this may be due to a disk getting too full, my "fullest" disk still has 182GB left available, so I don't think that's the problem.  I tried doing a reiserfsck and found one disk had a few errors, which were fixed by running with fix-fxable.  But even after that I still see the problem! 

Anyway, I'm at my wits end here.  I'd love it if someone could help me figure this one out! 

Oh, I just updated to unraid 6.1.9 today and am still seeing the problem.  I've attached the last syslog showing the Error, hopefully someone can help me get my server back up and working fully!

 

thanks,

Joe

unraid_crash.txt

Sounds like symptoms many of us experienced when moving to 6.x. It seems the newer software is less tolerant of reiserFS issues. Most of us just gave up troubleshooting the symptoms, and migrated all our disks to xfs. This is a slow process, as it involves moving data from disk to disk. You can't convert a reiserfs disk to xfs in place, you must format a blank target disk as xfs, and copy the data from your reiserfs disk to the blank xfs disk.

 

Long thread, but it has some good info on the process.

http://lime-technology.com/forum/index.php?topic=37490.0

Hmmm...well that sucks.  I still have the 2TB cache drive I've been using (just currently disabled).  Could I add this as a new disk, and use it to temporarily store the data, while I do the format?  I'm a little worried about moving data around since right now I can't seem to write anything to the server.

You could actually use it as a cache drive, formatted as xfs to start things off. One of the major symptoms is that disk reads from reiserfs disks are fine, just writes cause issues. So, if you format the new cache drive as xfs, and only write to it and see if the server acts ok as long as you don't write to a reiserfs drive, it's pretty much a sure bet there's your problem.

 

If you have a good handle on user and disk shares, the mover and cache drive interactions, you can use the cache drive to move data around to facilitate reformatting. If not, I suggest you read and digest the xfs conversion thread and ask questions before you do anything. It's pretty easy to lose data if you type something blindly without understanding what you are telling the system to do.

Hey that worked Jonathanm!

I used that disk as a new cache drive and created it using XFS.  After formatting and what not, mover is now working again and actually writing data to the disks!  Phew .  I thought I was screwed there for a while.  Thanks for the help!  Now I guess I'll need to start migrating all disks to XFS.  Not looking forward to that.