SCSI

Resetting/Rescanning

Say you've got a tape drive and it wanders off into oblivion (as mine seem to do quite often), this might work.

  1. find the wacky 4 number code that denotes the device in question. The wacky number is the host, channel, id and lun.

Either look for a line like this:

st 1:0:6:0: Attached scsi tape st0

in the output from dmesg.

Or look in /proc/scsi/scsi for the host, channel, id and lun.

Host: scsi1 Channel: 00 Id: 06 Lun: 00
  Vendor: HP       Model: Ultrium 1-SCSI   Rev: E21V
  Type:   Sequential-Access                ANSI SCSI revision: 03

The magic number we want is 1 0 6 0.

  1. remove the device from the kernel like this...

echo "scsi remove-single-device 1 0 6 0" > /proc/scsi/scsi
  1. re-add the device to the kernel like this...

echo "scsi add-single-device 1 0 6 0" > /proc/scsi/scsi


Check Consistency

I was having a lot of trouble with Zephir. New kernels seemed to hate the Megaraid drive. I tried booting to the raid card interface and running the consistency check. This took a long time, but after it was over it said "Corrected" and Zephir stopped complaining. CONCLUSION! Run that feature every now and again...


CategoryNotes

SCSI (last edited 2008-02-27 01:17:56 by localhost)