Hi

 

I've been using my unRAID server happily for years, until last week it got accidentally unplugged from the power.  When I plugged it back in and booted up I found the web page (e.g. http://tower) now runs painfully slow, I've timed it and it takes 3 minutes to switch between the Main/Shares/Users/Settings/Utils tabs!  And the parity check is running at a painful 183.28 KB/sec!  At this rate the parity check is going to take 208 days!

 

I've not changed any hardware or settings, and the other services (twonky, plex, mysql) I run on my unRAID box seem unaffected.  Copying files to and from it also seems unaffected.

 

I am running 5.0-rc12a

 

Does anyone have any ideas what is going on?  And how I can fix it?

 

Thanks

Kevin

Hi

 

I've been using my unRAID server happily for years, until last week it got accidentally unplugged from the power.  When I plugged it back in and booted up I found the web page (e.g. http://tower) now runs painfully slow, I've timed it and it takes 3 minutes to switch between the Main/Shares/Users/Settings/Utils tabs!  And the parity check is running at a painful 183.28 KB/sec!  At this rate the parity check is going to take 208 days!

 

I've not changed any hardware or settings, and the other services (twonky, plex, mysql) I run on my unRAID box seem unaffected.  Copying files to and from it also seems unaffected.

 

I am running 5.0-rc12a

 

Does anyone have any ideas what is going on?  And how I can fix it?

 

Thanks

Kevin

You might try looking in your syslog for clues.  Have you tried rebooting your unRAID server?

At this rate the parity check is going to take 208 days!

How long did your last parity check take?

I have tried rebooting, but it stays painfully slow.  It used to just take 6 or 7 hours to run the parity check, I used to leave it over night and it would always be done by morning.

 

The syslog is full of errors like this:

 

Jun 17 20:20:47 Tower kernel: ata2: SATA link up 1.5 Gbps (SStatus 113 SControl 310)

Jun 17 20:20:47 Tower kernel: ata2.00: configured for UDMA/33

Jun 17 20:20:47 Tower kernel: ata2: EH complete

Jun 17 20:20:47 Tower kernel: ata2.00: exception Emask 0x0 SAct 0x0 SErr 0x280000 action 0x0

Jun 17 20:20:47 Tower kernel: ata2.00: BMDMA2 stat 0x6d0009

Jun 17 20:20:47 Tower kernel: ata2: SError: { 10B8B BadCRC }

Jun 17 20:20:47 Tower kernel: ata2.00: failed command: READ DMA EXT

Jun 17 20:20:47 Tower kernel: ata2.00: cmd 25/00:00:bf:eb:d7/00:04:01:00:00/e0 tag 0 dma 524288 in

Jun 17 20:20:47 Tower kernel:          res 51/04:bf:00:ef:d7/00:00:01:00:00/e0 Emask 0x1 (device error)

Jun 17 20:20:47 Tower kernel: ata2.00: status: { DRDY ERR }

Jun 17 20:20:47 Tower kernel: ata2.00: error: { ABRT }

Jun 17 20:20:47 Tower kernel: ata2.00: failed to IDENTIFY (I/O error, err_mask=0x1)

Jun 17 20:20:47 Tower kernel: ata2.00: revalidation failed (errno=-5)

Jun 17 20:20:47 Tower kernel: ata2: hard resetting link

But I'm not sure what that means or what to do about it

 

Thanks!

Kevin

Could be better you post full syslog (Utils -> System Log)... if all errors refer to "ata2" device I guess it may be a problem with that hdd or data cable (you can identify what one by searching on syslog for ata2, first lines found, from system boot, should help you identify it).

 

Posting smart status for your hdd's may also be useful.

The whole syslog is 6mb, so I can't attach it all, but I have attached the first 2000 lines which is hopefully enough.

 

What is the best why to get the smart status of my hdds to share on here?

 

Thanks!

Kevin

syslog.head.txt

To get smart status just telnet your unraid machine and use:

smartctl -a /dev/sda

smartctl -a /dev/sdb

smartctl -a /dev/sdc

...

I've attached the smart status output for each drive.  Hope that helps!

 

Thanks!

Kevin

smartstatus.zip

Apparently the problem is with your 'sdb' hdd, though I'm not fully sure if these 'read dma' errors may be caused by communication problems with the hdd (data cable, controller, etc...) or an internal error on the hdd, you may try checking data cable anyway or just wait some expert to comment this.

Could be a loose or bad SATA cable to sdb. In any case, sdb also has pending sectors and needs to be rebuilt. If you have a spare use it now, then pre-clear sdb and check the number of current_pendong_sectors. The RAW value must be zero. If you don't have a spare then you can pre-clear sdb and then rebuild.

That fixed it!  Thanks nars and dgaschk!

 

I opened it up, disconnected and reconnected the sdb cables and now it is all working perfectly again!  My parity check is now running at a much more respectable 76.93MB/sec

 

I've ordered some replacement cables just incase.

 

Thanks!

Kevin

 

Does sdb still have pending sectors? Parity protection requires that all drives have zero pending sectors.