Ever since a recent power failure (may not be related) I've not been able to complete a parity check.  The server is also becoming unresponsive after a few hours.

 

I've had a similar problem in the past when using beta 14 which I believe was related to that spindown problem, but I'm not 100% sure on that. I've since reverted to beta 12 which seems to have resolved it, but now it seems to be happening again  

 

All drives are green in the webUI.  All the data seems to be intact.  Complete syslog attached.  Running beta 12a.

 

Thanks!

syslog.txt

It happened again after a reboot and without even starting the array.

Hmm... something very strange is now happening.  One of my disks came up as failed (the newest one I just replaced, interestingly enough).

I stopped the array and it said that the failed disk has been replaced and was now blue (it wasn't).  It thought the "new" disk was 2TB, when in fact it is only 1.5 TB.  I rebooted again and now it's expecting a 2 TB disk in place of it.

 

This is really bizarre.  Is it possible that hot swap bay is bad?

 

Checked and re-seated all the connections.  Reset the config and am rebuilding the parity.  Hope this works.

Parity keeps freezing during rebuild.  (Web UI unresponsive, no disk activity, shares are inaccessible)

 

Red drive hasn't come back (... yet)

 

Any ideas?

Doing a memtest now.  Seems to be okay.

Mem test checked out fine, it's still happening though.  It seems my problem is the "BLK_EH_NOT_HANDLED" error.

Latest syslog attached.  What does this error mean?

syslog.txt

Reading around the forums it seems like it's a mystery with the 2xSupermicro AOC-SASLP-MV8 SAS controllers.  I did another hard reboot after another freeze while rebuilding the parity, stopped the rebuild, spun down all the disks and spun them back up without issue.  I then disabled spindown completely (used to be 2 hours) and am starting another rebuild.

 

I'm really just guessing over here as I'm not intimately familiar with linux enough to be comfortable to really poke around.

 

Any feedback would be much appreciated.  Thanks guys!

Big surprise it froze again.  It seemed to have gotten farther this time, however.  Latest syslog attached.  Same error I believe.

 

Anyone?  Bueller?  Bueller?

syslog.txt

Bump.

This is a known bug. Please read the announcements thread.

Yeah, I just went through the whole thread and saw that.  Not sure what the prescribed course of action is, though.  I am going to try beta 14 right now.  If not, should I go back to 6a?

Yes, I believe you have to go backwards to something like 6a.

Okay, I'll try that.  Am I making this up or do I remember having to run some permission script between 6a and 12a?  Would that have messed things up if I try to go back?

Beta 6a isn't even available anywhere...  

 

Trying again with beta 11

That worked!

 

So for anyone wondering, if you're getting the "BLK_EH_NOT_HANDLED" error in 5.0, downgrade to beta 11.

Has this been solved - or do we still need to downgrade?

Has this been solved - or do we still need to downgrade?

Any reason you are not trying 5.0-rc8a?  As far as I know, it has all the fixes.

 

Joe L.

Thanks Joe - I will try that tonight!