Datalink did actually call back and had me try a few other things but nothing 
they did helped.

Prior to calling them I'd been thinking about use of sg_reset which is 
something we do when adding SAN disks.  Essentially "sg_reset -h <device>" 
resets the "host" (HBA) of the device in question.   In desperation I tried 
doing this on the tape device "sg_reset /dev/st0" but that returned an I/O 
error which tended to confirm that somehow the devices though showing up in 
/proc/scsi/scsi and lsscsi had somehow lost connection with the hardware.

Due to this I ran the Qlogic provided utility that resets the Qlogic cards:
ql-dynamic-tgt-lun-disc.sh
This warns not to run if there is data but since we have two HBAs and past 
experience shows it does the reset one HBA at a time it didn't seem likely to 
cause a problem.   (However I did test it on a dev system similarly configured 
before running it on the production system.)
Although it said it found no device changes it was clear that it did find the 
devices because after doing that trying the sg_reset worked without giving the 
I/O error.  (It did say there was no medium but I expected that since I'd chose 
an empty tape drive.)   For good measure I did the sg_reset on a device on each 
of the fibre cards.  However, I'm not sure this was actually required.

After the resets I ran /usr/openv/volmgr/bin/scan again and it now showed the 
tape drives where before it had only output header type information.

Once that was done I used the NBU Java GUI Storage Device Configuration and 
deselected all hosts except for the one that had been the issue.   I accepted 
defaults on all screens except the one for storage unit configuration (we have 
our own naming convention for storage units) where I deselected what it gave.   

After doing that I verified our original storage unit names for the above hosts 
were still in place then ran two test backups to tape successfully.

Thanks to all who tried to help.

-----Original Message-----
From: [email protected] 
[mailto:[email protected]] On Behalf Of Lightner, Jeff
Sent: Thursday, January 12, 2012 2:13 PM
To: [email protected]
Subject: Re: [Veritas-bu] MISSING PATH for tape devices in RHEL5 for NetBackup 
7.1

The full lsscsi output shows disks etc... so I've stripped it to only show the 
tape stuff:
lsscsi |egrep -i "quantum|tape|hp|st"
[0:0:2:0]    storage QUANTUM  Scalar i2000     5A87  -
[0:0:2:1]    tape    HP       Ultrium 3-SCSI   M69Z  /dev/st0
[0:0:2:2]    tape    HP       Ultrium 3-SCSI   M69Z  /dev/st1
[0:0:2:3]    tape    HP       Ultrium 5-SCSI   I3AZ  /dev/st2
[0:0:2:4]    tape    HP       Ultrium 5-SCSI   I3AZ  /dev/st3
[0:0:3:0]    storage QUANTUM  Scalar i2000     5A87  -
[0:0:3:1]    tape    HP       Ultrium 5-SCSI   I3AZ  /dev/st4
[0:0:3:2]    tape    HP       Ultrium 5-SCSI   I3AZ  /dev/st5
[0:0:3:3]    tape    HP       Ultrium 5-SCSI   I3AZ  /dev/st6
[0:0:3:4]    tape    HP       Ultrium 5-SCSI   I3AZ  /dev/st7
[1:0:1:73]   disk    HITACHI  OPEN-V           7002  /dev/sdhp
[1:0:2:0]    storage QUANTUM  Scalar i2000     5A87  -
[1:0:2:1]    tape    HP       Ultrium 5-SCSI   I3AZ  /dev/st8
[1:0:2:2]    tape    HP       Ultrium 5-SCSI   I3AZ  /dev/st9
[1:0:2:3]    tape    HP       Ultrium 5-SCSI   I3AZ  /dev/st10
[1:0:2:4]    tape    HP       Ultrium 5-SCSI   I3AZ  /dev/st11

Thanks for suggesting lsscsi - I'd forgotten about it.   The nice thing about 
that is it equates the path seen in /proc/scsi/scsi to the device path in /dev.

Running an mt -f /dev/nst0 status on this host gives an I/O error
mt -f /dev/nst0 status
/dev/nst0: Input/output error

Whereas on another media server just like the above the output is different:
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







-----Original Message-----
From: [email protected] 
[mailto:[email protected]] On Behalf Of Bluejay Adametz
Sent: Thursday, January 12, 2012 1:31 PM
To: [email protected]
Subject: Re: [Veritas-bu] MISSING PATH for tape devices in RHEL5 for NetBackup 
7.1

> However, if I do a cat /proc/scsi/scsi I see the 3 paths to the library
> (Quantum i6000) blades and the 12 HP Ultrium (10 Ultrium 5 and 2 Ultirum 3).

Try using lsscsi (from the lsscsi package, if you need to install it)
and see that all your tape drives show up with the device names you
expect.

If they're there, you can use the mt utility (from mt-st) to see if
you can actually access them: the status option should return some
information from the drive.

If all that works, themn the next suspect would be NetBackup.

                                                         - Bluejay Adametz

Most quarrels amplify a misunderstanding. - Andre Gide
NOTICE: This message, including any attachments, is only for the use of
the intended recipient(s) and may contain confidential and privileged
information, or information otherwise protected from disclosure by law.
If the reader of this message is not the intended recipient, you are
hereby notified that any use, disclosure, copying, dissemination or
distribution of this message or any of its attachments is strictly
prohibited.  If you received this message in error, please contact the
sender immediately by reply email and destroy this message, including all
attachments, and any copies thereof.

_______________________________________________
Veritas-bu maillist  -  [email protected]
http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu




Athena(r), Created for the Cause(tm)
Making a Difference in the Fight Against Breast Cancer

---------------------------------
CONFIDENTIALITY NOTICE: This e-mail may contain privileged or confidential 
information and is for the sole use of the intended recipient(s). If you are 
not the intended recipient, any disclosure, copying, distribution, or use of 
the contents of this information is prohibited and may be unlawful. If you have 
received this electronic transmission in error, please reply immediately to the 
sender that you have received the message in error, and delete it. Thank you.
----------------------------------

_______________________________________________
Veritas-bu maillist  -  [email protected]
http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
_______________________________________________
Veritas-bu maillist  -  [email protected]
http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu

Reply via email to