SUMMARY: SDLT320 tape full

From: Bolay, Vincent <vincent.bolay_at_medtronic.com>
Date: Fri Jun 04 2004 - 11:04:50 EDT
Thanks to those who answered:

Tim Evans
Christopher Dantos
Laurence Moughan
Jason Zawacki

It seems that it's a known problem. Here is Jason's answer:
-----------------
I had that blocksize error but the entry in st.conf fixed that.  I also had
the tapes reading as full when they weren't - and I fixed it by doing four
different things (maybe all three were needed, or just one of them, I'm not
sure but I don't get that error anymore).

I have a Quantum M1800 tape library, with 4 SDLT320 drives.

1) I had all four drives on one SCSI3 interface, which is too slow to push
data to all 4 drives.  So, I ordered 2 X6758A cards (dual channel SCSI3, and
gave each drive its own channel.)
2) I may have actually had a bad drive, and got that replaced.
3) the st.conf entry you note below.
4) Had Quantum send someone to upgrade the firmware on each drive

It's important to note that even though the tape is marked full and isn't, I
wasn't able to mark it "appendable" in Solstice Backup and have it write to
the tape again.  It just wouldn't.

I also had problems recovering from those sorts of tapes.  I ended up
relabelling them all and then they worked normally.

We're also running EBS 7.1.1 instead of 7.0 - on a Solaris 9 machine - a
SunFire v240.

Now it's been running for a few months now and the only problem is every now
and then we get unmount timeouts, but they fix themselves and have never
caused a backup or recover to fail (it just needs to wait until it times
out, then it works fine.)
----------------

Right now I upgraded to EBS 7.1 with patch 116826-2 (Build 277). I also
patched with the sd/st patch 113277-21.

I will post an updated if it solves my issue.

Thanks
Vincent
>  -----Original Message-----
> From: 	Bolay, Vincent
> Sent:	Thursday, June 03, 2004 4:54 PM
> To:	'sunmanagers@sunmanagers.org'
> Subject:	SDLT320 tape full
>
> Hello,
>
> Sometimes I get an error message telling that a tape is full, although
that's not the case.
> This is my configuration:
>
> Solaris 9
> EBS 7.0 Build 322 (Enterprise Backup Software, previously Solstice Backup)
> 1 Storagetek L80 library with 2 Quantum SDLT320 drives attached to a 280R
(server)
> 1 Storagetek L20 library with 2 Quantum SDLT320 drives attached to a 220R
(storage node)
> SuperDLTtape I (minimum capacity 160 GB)
>
> Extract of /nsr/logs/daemon.log:
> -----------
> 05/30/04 03:59:23 nsrd: media notice: sdlt320 tape WCX015 on
rd=lau-backup-b:/dev/rmt/1ubn is full
> 05/30/04 03:59:23 nsrd: media notice: sdlt320 tape WCX015 used 40 GB of 160
GB capacity
> 05/30/04 03:59:48 nsrd: media notice: Volume "WCX015" on device
"rd=lau-backup-b:/dev/rmt/1ubn": Block size is 32768 bytes not 131072 bytes.
Verify the device configuration. Tape positioning by record is disabled.
> 05/30/04 04:02:38 nsrd: media warning: rd=lau-backup-b:/dev/rmt/1ubn
reading: fsr 7218 read: I/O error
> 05/30/04 04:02:38 nsrd: media emergency: could not position WCX015 to file
22, record 7220
> 05/30/04 04:02:38 nsrd: media warning: rd=lau-backup-b:/dev/rmt/1ubn
reading: I/O error
> 05/30/04 04:02:38 nsrd: media warning: rd=lau-backup-b:/dev/rmt/1ubn
reading: I/O error
> 05/30/04 04:02:38 nsrd: media warning: rd=lau-backup-b:/dev/rmt/1ubn
reading: I/O error
> 05/30/04 04:02:38 nsrd: media warning: rd=lau-backup-b:/dev/rmt/1ubn
reading: I/O error
> 05/30/04 04:02:38 nsrd: media warning: rd=lau-backup-b:/dev/rmt/1ubn
reading: I/O error
> 05/30/04 04:02:38 nsrd: media warning: rd=lau-backup-b:/dev/rmt/1ubn
reading: I/O error
> 05/30/04 04:02:38 nsrd: media warning: verification of volume "WCX015",
volid 3954395772 failed, can not read record 7220 of file 22 on sdlt320 tape
WCX015
> 05/30/04 04:02:38 nsrd: media notice: verification of volume "WCX015", volid
3954395772 failed, volume is being marked as full.
> -----------
> Extract of /var/adm/messages:>
> -----------
> May 30 03:59:23 lau-backup-b scsi: [ID 107833 kern.warning] WARNING:
/pci@1f,4000/pci@4/scsi@4/st@1,0 (st50):
> May 30 03:59:23 lau-backup-b    Error for Command: write
Error Level: Fatal
> May 30 03:59:23 lau-backup-b scsi: [ID 107833 kern.notice]      Requested
Block: 7222                      Error Block: 7222
> May 30 03:59:23 lau-backup-b scsi: [ID 107833 kern.notice]      Vendor:
QUANTUM                           Serial Number:  F    i
> May 30 03:59:23 lau-backup-b scsi: [ID 107833 kern.notice]      Sense Key:
Media Error
> May 30 03:59:23 lau-backup-b scsi: [ID 107833 kern.notice]      ASC: 0xc
(write error), ASCQ: 0x0, FRU: 0x0
> May 30 04:02:38 lau-backup-b scsi: [ID 107833 kern.warning] WARNING:
/pci@1f,4000/pci@4/scsi@4/st@1,0 (st50):
> May 30 04:02:38 lau-backup-b    SCSI transport failed: reason 'timeout':
giving up
> -----------
> The blocksize is set up to 128 kB.
>
> According to our local Sun support we added the following lines in
/kernel/drv/st.conf:
>
> tape-config-list=
> "QUANTUM SDLT320",        "Quantum Super DLT 320",            "SDLT320";
>
> SDLT320       = 2, 0x38,    0, 0x8649,  4, 0x90, 0x91, 0x92, 0x93, 3, 120,
120,
> 3600, 3600, 3600, 3600, 21600;
>
> I get the same error with or without this addition.
>
> Is anyone experiencing the same problem?
>
> I will summarize.
> Vincent
>
> Vincent Bolay
> Systems Engineer
> Medtronic Europe S`rl
> Route du Molliau 31
> Case postale
> CH-1131 Tolochenaz
> Switzerland
> Phone: (+41 -21) 802 75 34
> Fax: (+41 -21) 802 79 11
> E-mail: vincent.bolay@medtronic.com
_______________________________________________
sunmanagers mailing list
sunmanagers@sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
Received on Fri Jun 4 11:04:42 2004

This archive was generated by hypermail 2.1.8 : Thu Mar 03 2016 - 06:43:33 EST