Re: [Networker] How to get around block size errors?

Sorry, I'm not a Linux guy. Might want to look at the docs for the
driver for your LSI cards. Also, might want to see if there is a newer
driver and/or LSI firmware. I just know that the block size error
correlated in our case to a underlying SCSI problem including cable
issues, drive issues (needed more cleaning), and some media issues. What
happens is that the driver requests to read a 64k block, and only gets
512bytes back. If you can get SCSI error reporting turned on, you will
see the SCSI error happening at the same time as the Legato error.


These type of errors were one reason we migrated our STKL180 to native
FC. SCSI has too much voodoo.



--
Matthew Huff           | One Manhattanville Rd
Director of Operations | Purchase, NY 10577
OTA LLC                | Phone: 914-460-4039 
mailto:mhuff AT ox DOT com    | Fax:   914-460-4139   
-----Original Message-----
From: Legato NetWorker discussion [mailto:NETWORKER AT LISTSERV.TEMPLE DOT EDU]
On Behalf Of George Sinclair
Sent: Thursday, November 17, 2005 8:30 PM
To: NETWORKER AT LISTSERV.TEMPLE DOT EDU
Subject: Re: [Networker] How to get around block size errors?

We're using three dual channel Ultra320 PCI LSI logic SCSI adapter cards
(LSI22320-R) , each on its own bus.

STL L80 library (4 LTO 1 drives):
drives 1-2 use channel A and B on card 1. Picker is daisy chained to
drive 1.
drives 3-4 use channel A and B on card 2.

Quantum P1000 (2 SDLT 1 drives)
drive 1 uses channel A on card 3. Picker is daisy chanined to drive 1.
drive 2 uses chaneel B on card 3.

How do I turn max error logging on?

Anyone using these same cards on Linux RedHat? Is there a trick to
configure them correctly?

George

Matthew Huff wrote:

> 
>When I've seen that error, usually I've traced it down to an actual 
>SCSI error. Check your messages log file, and make sure that your st 
>driver has max error logging on. What adapter are you using on the 
>storage node?
>
>--
>Matthew Huff           | One Manhattanville Rd
>Director of Operations | Purchase, NY 10577
>OTA LLC                | Phone: 914-460-4039 
>mailto:mhuff AT ox DOT com    | Fax:   914-460-4139   
>-----Original Message-----
>From: Legato NetWorker discussion 
>[mailto:NETWORKER AT LISTSERV.TEMPLE DOT EDU]
>On Behalf Of George Sinclair
>Sent: Thursday, November 17, 2005 7:44 PM
>To: NETWORKER AT LISTSERV.TEMPLE DOT EDU
>Subject: [Networker] How to get around block size errors?
>
>Is there any recommended way to recover data when you get a block error

>during nwrecover?
>
>nsrd: media notice: Volume "volname" on device
>"rd=storagenode:/dev/nst2": Block size is 512 bytes not 65536 bytes.
>Verify the device configuration. Tape positioning by record is
disabled.
>
>At that point, the recovery just hangs as the tape sits idle. We see 
>this a lot. Sometimes, the error occurs during backups and sometimes 
>during recoveries, but not always. Seems random., but seems to occur 
>about 40% of the time we need to recover data. Yikes!!!
>
>Whenever I've scanned one of these tapes, to see the actual block size,

>the reported block size is always what NetWorker *wants* or is 
>expecting, not what it complained about. The LTO 1 tapes are always 64 
>KB and the SDLT 1 tapes are always 128 KB. Also, checking the devices 
>under the GUI shows these respective sizes. Not sure why NetWorker is 
>getting confused.
>
>Is there anything that can be done to at least temporarily allow you to

>get around this problem? Anything to provide relief other than to pack 
>up and move away to a far off land with no supper?
>
>We were told by Legato tech support (we sent them our log files) that 
>engineering said there were issues that were fixed in later releases, 
>but no clearer answers beyond that -- nothing specific regarding the 
>actual block problems/errors themselves and how to resolve them. We 
>have received a new media kit (7.2.1) and plan to migrate to a new 
>server to run this, but we can't do it for a few more weeks. Has anyone

>seen the same problems go away after upgrading to a more recent 
>release? I know there have been a lot of reported block problems with
Windows OS.
>
>Currently, we're running an old release 6.1.1 on Solaris 2.8. but the 
>devices we're using are managed by a single RedHat Linux storage node 
>(6.1.1). The devices are SDLT 1 and LTO 1 drives. The storage node has 
>2 libraries, one with LTO drives, one with SDLT. We've seen these 
>problems on all the devices, though, and I think this problem has been 
>persistent for a long time. We use LSI logic SCSI cards. We're using 
>Fuji and Maxell tapes.
>
>We have what we think is a properly configured stinit.def file (see 
>below), but if there's anything we can do in the mean time to provide 
>some relief, with minimal impact, or anything we can try, that would be

>nice. I hope these problems go away with the new release. I wonder, 
>though, about accessing tapes that were labeled and written with our 
>current release?
>
># Seagate Ultrium LTO
>manufacturer=SEAGATE model = "ULTRIUM06242-XXX" {
>scsi2logical=1 can-bsr auto-lock
>mode1 blocksize=0
>}
>
># SDLT220
>manufacturer="QUANTUM" model = "SuperDLT1" {
>scsi2logical=1
>can-bsr=1
>auto-lock=0
>two-fms=0
>drive-buffering=1
>buffer-writes
>read-ahead=1
>async-writes=1
>can-partitions=0
>fast-mteom=1
>#
># If your stinit supports the timeouts:
>timeout=3600 # 1 hour
>long-timeout=14400 # 4 hours
>#
>mode1 blocksize=0 density=0x48 compression=1    # 110 GB + compression
>mode2 blocksize=0 density=0x48 compression=0    # 110 GB, no
compression
>}
>
>
>Thanks in advance.
>
>George
>
>To sign off this list, send email to listserv AT listserv.temple DOT edu and 
>type "signoff networker" in the body of the email. Please write to 
>networker-request AT listserv.temple DOT edu if you have any problems wit this

>list. You can access the archives at 
>http://listserv.temple.edu/archives/networker.html or via RSS at 
>http://listserv.temple.edu/cgi-bin/wa?RSS&L=NETWORKER
>
>  
>

To sign off this list, send email to listserv AT listserv.temple DOT edu and
type "signoff networker" in the body of the email. Please write to
networker-request AT listserv.temple DOT edu if you have any problems wit this
list. You can access the archives at
http://listserv.temple.edu/archives/networker.html or via RSS at
http://listserv.temple.edu/cgi-bin/wa?RSS&L=NETWORKER

To sign off this list, send email to listserv AT listserv.temple DOT edu and 
type "signoff networker" in the
body of the email. Please write to networker-request AT listserv.temple DOT edu 
if you have any problems
wit this list. You can access the archives at 
http://listserv.temple.edu/archives/networker.html or
via RSS at http://listserv.temple.edu/cgi-bin/wa?RSS&L=NETWORKER