[SOLVED] Exabyte vxa 3 tape drive - mt -f /dev/nst0 no such address or device

davide123 · 01-03-2011, 02:19 PM

I have an LTO-3 tape drive. It shows

Code:

cat /proc/scsi/scsi 
Host: scsi0 Channel: 01 Id: 03 Lun: 00
  Vendor: EXABYTE  Model: VXA-3            Rev: 3233
  Type:   Sequential-Access                ANSI SCSI revision: 03
Host: scsi0 Channel: 02 Id: 08 Lun: 00

when I do

Code:

mt -f /dev/st0 status
/dev/st0: No such device or address

Code:

mt -f /dev/nst0 status
/dev/nst0: No such device or address

I am running as root
Would like to NOT reboot this server. (WEBSITE) It is running Red Hat Enterprise Linux ES release 4 (Nahant Update 8)
Thank you!

stress_junkie · 01-03-2011, 09:15 PM

Evidently Linux mapped the drive to a different device. Use less to scan /var/log/messages for information.

Code:

less /var/log/messages

Search for EXABYTE or TAPE as you would do in vi.

That will tell you how Linux mapped the drive to a device. Here is an example of how Linux maps my ATA DVDROM to /dev/sr0

Quote:

Dec 29 16:16:13 f14 kernel: [ 0.621366] scsi 0:0:0:0: CD-ROM DVDRO
M PATA 16X 1A31 PQ: 0 ANSI: 5
Dec 29 16:16:13 f14 kernel: [ 0.625530] sr0: scsi3-mmc drive: 48x/48x cd/rw x
a/form2 cdda tray
Dec 29 16:16:13 f14 kernel: [ 0.625534] Uniform CD-ROM driver Revision: 3.20

davide123 · 01-04-2011, 07:49 AM

Hi thanks for the reply. I do not see anything in messages relating to my tape drive. In dmesg i see the following:

Code:

[root@**** log]# dmesg |grep EXA
  Vendor: EXABYTE   Model: VXA-3             Rev: 3233
[root@***** log]# dmesg |grep tape
Attached scsi tape st0 at scsi0, channel 1, id 3, lun 0
st0: Write not multiple of tape block size.
st0: Write not multiple of tape block size.

also here is a bit more info:

Code:

[root@**** log]# ls -l /dev/tape
lrwxrwxrwx  1 root root 4 Dec  2 09:43 /dev/tape -> nst0

[root@**** log]# ls -l /dev/nst0
crw-rw----  1 root disk 9, 128 Dec  2 09:43 /dev/nst0

Code:

[root@**** ~]# dmesg | grep -i scsi
SCSI subsystem initialized
scsi0 : IBM PCI ServeRAID 7.12.05  Build 761
  Type:   Direct-Access                      ANSI SCSI revision: 02
SCSI device sda: 143372288 512-byte hdwr sectors (73407 MB)
SCSI device sda: 143372288 512-byte hdwr sectors (73407 MB)
Attached scsi disk sda at scsi0, channel 0, id 0, lun 0
  Type:   Direct-Access                      ANSI SCSI revision: 02
SCSI device sdb: 573493248 512-byte hdwr sectors (293629 MB)
SCSI device sdb: 573493248 512-byte hdwr sectors (293629 MB)
Attached scsi disk sdb at scsi0, channel 0, id 1, lun 0
  Type:   Processor                          ANSI SCSI revision: 02
  Type:   Sequential-Access                  ANSI SCSI revision: 03
  Type:   Processor                          ANSI SCSI revision: 02
Attached scsi generic sg0 at scsi0, channel 0, id 0, lun 0,  type 0
Attached scsi generic sg1 at scsi0, channel 0, id 1, lun 0,  type 0
Attached scsi generic sg2 at scsi0, channel 0, id 15, lun 0,  type 3
Attached scsi generic sg3 at scsi0, channel 1, id 3, lun 0,  type 1
Attached scsi generic sg4 at scsi0, channel 2, id 8, lun 0,  type 3
Attached scsi tape st0 at scsi0, channel 1, id 3, lun 0
SCSI error : <0 0 0 0> return code = 0x70000
scsi1 : SCSI emulation for USB Mass Storage devices
  Type:   Direct-Access                      ANSI SCSI revision: 02
SCSI device sdc: 976773168 512-byte hdwr sectors (500108 MB)
SCSI device sdc: 976773168 512-byte hdwr sectors (500108 MB)
Attached scsi disk sdc at scsi1, channel 0, id 0, lun 0
Attached scsi generic sg5 at scsi1, channel 0, id 0, lun 0,  type 0
scsi: Device offlined - not ready after error recovery: host 0 channel 1 id 3 lun 0

stress_junkie · 01-04-2011, 11:59 AM

Quote:

Originally Posted by davide123

Code:

[root@**** log]# dmesg |grep EXA
  Vendor: EXABYTE   Model: VXA-3             Rev: 3233
[root@***** log]# dmesg |grep tape
Attached scsi tape st0 at scsi0, channel 1, id 3, lun 0
st0: Write not multiple of tape block size.
st0: Write not multiple of tape block size.

The first three lines certainly indicate that the tape drive is mapped to /dev/st0, and we know that /dev/nst0 is included as the non-rewinding mode.

The last two lines interest me. I wonder if they indicate that the tape drive is not working properly. Can you physically access the tape drive? Maybe if you remove the SCSI cable, clean the contacts, and reattach the cable we would learn something helpful in dmesg or in the /var/log/messages after the device is reattached. If possible clean the contacts on both ends of the cable.

Another thought. Is there a tape already loaded in the drive? If so then remove it. The tape may be bad. The error message may pertain to a bad tape cartridge. Try using a different tape cartridge.

davide123 · 01-04-2011, 12:37 PM

Well will re attaching the SCSI cable require a reboot? (I have replaced the cable with a brand new one straight out of the bag) I was under the impression the OS will only find the SCSI devices upon boot. I cannot re-boot at the time being. Hopefully tomorrow, because good 'ole reboot, in my opinion is one of the best troubleshooting options.

And no, there is no tape in the drive

stress_junkie · 01-04-2011, 05:39 PM

Quote:

Originally Posted by davide123

Well will re attaching the SCSI cable require a reboot? (I have replaced the cable with a brand new one straight out of the bag) I was under the impression the OS will only find the SCSI devices upon boot. I cannot re-boot at the time being.

I don't know about RHEL 4 but I believe that RHEL 5 and anything more modern will automatically detect devices when they are attached. If you have already changed the cable then don't bother with this.

Quote:

Originally Posted by davide123

Hopefully tomorrow, because good 'ole reboot, in my opinion is one of the best troubleshooting options.

You and me both.

Quote:

Originally Posted by davide123

And no, there is no tape in the drive

Okay, good, and bad. If only it were that simple.

That error message that I mentioned above may indicate that the tape drive is not working. Can you attach the tape drive to another machine to test it?

davide123 · 01-05-2011, 07:55 AM

I am going to try switching the tape drive to another machin but first I want to try this

Quote:

To rescan all other SCSI attached storage, a rescan should be issued:

shell> echo "- - -" > /sys/class/scsi_host/host#/scan

Replace the “#” with the number of the SCSI bus to be rescanned.

I show host0 & host1 on my machine. I have a RAID controller/card on my machine and I am worried about rescanning the wrong host#.... Do you know how I would figure out which host# I should issue this command with?

as per above I am guessing

Quote:

scsi: Device offlined - not ready after error recovery: host 0 channel 1 id 3 lun 0

UPDATE!!
Well scratch that I guess...
i tried:

Code:

[root@stpweb2 ~]# echo "- - -" > /sys/class/scsi_host/host0/scan
[root@**** ~]# mt -f /dev/nst0 status
/dev/nst0: No such device or address
[root@***** ~]# mt -f /dev/st0 status
/dev/st0: No such device or address
[root@***** ~]# echo "- - -" > /sys/class/scsi_host/host0/scan
[root@***** ~]# mt -f /dev/st0 status
/dev/st0: No such device or address
[root@***** ~]# echo "1" > /sys/class/scsi_host/host0/scan
[root@***** ~]# mt -f /dev/st0 status
/dev/st0: No such device or address
[root@**** ~]# echo "0" > /sys/class/scsi_host/host0/scan
[root@**** ~]# mt -f /dev/st0 status
/dev/st0: No such device or address
[root@***** ~]# echo "2" > /sys/class/scsi_host/host0/scan
[root@***** ~]# mt -f /dev/st0 status
/dev/st0: No such device or address
[root@**** ~]# echo "1 2 0" > /sys/class/scsi_host/host0/scan
[root@**** ~]# mt -f /dev/st0 status
/dev/st0: No such device or address

no joy.... I am just waiting for an ok to reboot the machine. I am thinking switching the "bad" tape drive while the "good" machine is on wont do any good, I am thinking I will have to reboot that as well. (Also RHEL 4)
So Once I get the ok to reboot the "bad" machine I will and hopefully it will pick up the tape drive.

davide123 · 01-05-2011, 09:58 AM

UPDATE #2!!
So I was able to re-boot the server. It picked up the Tape Drive.

Code:

[root@**** Linux]# mt -f /dev/nst0 status
SCSI 2 tape drive:
File number=-1, block number=-1, partition=0.
Tape block size 0 bytes. Density code 0x0 (default).
Soft error count since last status=0
General status bits on (50000):
 DR_OPEN IM_REP_EN

However, It will not take a tape. I put a tape in it blinks a green center light, then blinks the 2 outside amber lights and spits the tape back out.

Not Sure How To Get Around This?
I did put in a support ticket email to Tandberg, on their site the 2 amber lights indicate "unrecoverable error" I have tried power cycling the tape drive, different tapes, and reboot of the server all no joy....

stress_junkie · 01-07-2011, 04:17 PM

It sounds like the tape drive is broken.

davide123 · 01-10-2011, 04:49 AM

it does... this is what tandberg told me....

Quote:

Thank you for submitting your Technical Support email request. We have opened a new case for, Case ID 261693.

If this is SCSI connect please power off the drive and your Linux server and disconnect the VXA-3 from the bus and power up the drive. Insert a tape while the drive is disconnected, If the drive continues to eject the tape. The drive needs service. Please call Technical support at 303 417-7792 and reference the case # 261693 for assistance.

davide123 · 01-18-2011, 08:04 AM

What Tandberg told me solved the issue with this Tape Drive. As far as the error mt-f I/O error. I solved this by doing

Quote:

mt -f /dev/nst0 setblk 0

This sets tape block size automatic.