I just got done building a 12 drive array and shifting files around between the drives.  I then tried to remove one drive from the array and I have not been able to get unRAID to rebuild the parity drive since.  The server keeps freezing.  syslog shows the following:

 

Tower Kernel [xxx] can't srrink filesystem on-line

Tower kernel: [  62.194065] hdj: drive_cmd: status =0x51 { DriveReady SeekComplete Error }

Tower kernel: [  62194071] hdj: drive_cmd: error=0x04 { DriveStatusError

Tower kernel: [  62.194074] ide failed opcode was: 0xb0

Tower emhttp[1512]: get_temperature: ioctl (smart): Input/output error

Tower emhttp[1369]: shcmd (18): killall -HUP smdb

Tower login[1370]: ROOT LOGIN on 'tty1'

 

Any ideas?  I can't keep the server up and running long enough to do any troubleshooting.  The only way I was able to get this output was tailing syslog and then typing it manually here.

You can look at the Devices page to see which disk is assigned as "hdj" and then check that disk's cabling, power, and master/slave/cable-select jumper.  If still no-go then try unassigning that disk on Devices page.  If still no-go then uncable or remove the disk.

 

An easier way to capture the syslog would be to open telnet window on your PC, and use the command:

 

tail -f /var/log/syslog

 

Then you can select/copy/paste the contents.

 

Are you using version 4.2.1?

I am running 4.2.1.  I no longer think drive hdj is the issue.  Once restarted I received an error each time I jumpted back and forth between devices and the main page .  The last error in syslog was not the error I received when the server hung up.  In fact there were no errors when the system hung up.

 

 

Dec 27 06:24:37 Tower kernel: [  182.049127] hdj: drive_cmd: status=0x51 { Drive

Ready SeekComplete Error }

Dec 27 06:24:37 Tower kernel: [  182.049133] hdj: drive_cmd: error=0x04 { DriveS

tatusError }

Dec 27 06:24:37 Tower kernel: [  182.049136] ide: failed opcode was: 0xb0

Dec 27 06:24:37 Tower emhttp[1583]: get_temperature: ioctl (smart): Input/output

error

Dec 27 06:24:48 Tower kernel: [  192.461482] hdj: drive_cmd: status=0x51 { Drive

Ready SeekComplete Error }

Dec 27 06:24:48 Tower kernel: [  192.461488] hdj: drive_cmd: error=0x04 { DriveS

tatusError }

Dec 27 06:24:48 Tower kernel: [  192.461491] ide: failed opcode was: 0xb0

Dec 27 06:24:48 Tower emhttp[1594]: get_temperature: ioctl (smart): Input/output

error

Dec 27 06:24:57 Tower kernel: [  201.876712] hdj: drive_cmd: status=0x51 { Drive

Ready SeekComplete Error }

Dec 27 06:24:57 Tower kernel: [  201.876718] hdj: drive_cmd: error=0x04 { DriveS

tatusError }

Dec 27 06:24:57 Tower kernel: [  201.876721] ide: failed opcode was: 0xb0

Dec 27 06:24:57 Tower emhttp[1605]: get_temperature: ioctl (smart): Input/output

error

 

I've tried disconnecting hdj and hdi and restart without them in the array and I the server still hangs up.  I've replaced memory, removed additional drives and controllers.  Nothing seems to keep the server up. 

 

Everything had been running for weeks on 3 drives and 4.2.1 basic.  The last two days I've been on the server pro version and was running fine until I removed the first drive.

I think I've narrowed it down to a specific controller card. I moved two of four drives to another controller and I've not had a reboot since.  I'll live without the extra two drives until I can find a working controller.  Any chance I can use a promise SX4000?  It is the only extra IDE controller I have here.

 

Brian

 

Is that a straight IDE expansion card or one of their RAID models? If it's RAId I woldn't think so but hey try it and see.

Did you ever determine if that SX4000 worked? I have one too.

 

BTW...yes, it is a raid controller. I don't know if it is required that you set up raid on it.