scsi time outs and sense data errors w/ tape drive

Hubbard, David dhubbard at dino.hostasaurus.com
Tue Jun 12 10:47:00 CDT 2001


Most tape drives that are capable of hardware compression
will reset to using hardware compression each time they
are reset or power cycled.  When you moved it to the new
box did you use the same cable?  And possibly the same
terminator?  Usually timeouts are caused by improper
cables or termination as well as other devices on the
bus.  For example, I cannot reliably burn CD's on my linux
pc at home if I have my jaz drive on the same scsi chain.

But anyway, with this problem, I'd check your termination
and cable.  Make sure you're not double terminating too,
i.e. the tape drive's switch is set for termination and
you've got a terminator hanging off the end of your scsi
cable.  If you swap the cable and ensure the termination
is ok, you might want to get the drive replaced.  I had
a PowerVault 120T 8-tape auto-loader, based on a Sony
DDS4 drive, have writing problems and Dell sent me a 
replacement, things worked fine after that.  I'm still
having trouble finding good linux backup software but
that's another story... :-)

Dave

-----Original Message-----
From: Chris Harwell [mailto:charwell at digitalpulp.com]
Sent: Tuesday, June 12, 2001 12:44 PM
To: linux-poweredge at dell.com
Subject: scsi time outs and sense data errors w/ tape drive 


greetings,

i've been seeing scsi time outs followed by
sense data errors with a dell poweredge and sony tape drive (more specs
below).

i've actually moved the tape drive to a different box (redhat 6.2 to 7.1)
and changed the cable but i get the same problem... which i only see
periodically,
seemingly when a lot of data may be going to the tape drive and perhaps
overwhelms it.

any suggestions? is there an FAQ i should look at?

do you know how to increase the timeout period or buffering?

i had fooled myself into believing the scsi resets had gone away
when i turned off datcompression ( su - -c "mt -f /dev/nst0 datcompression
0" )
 that had been on for some strange reason, but it didn't - though i did
notice it
claimed to be back on after the problem manifested again.

thanks :>

-- 
chris
charwell at digitalpulp.com



i'm using the dump, tar and mt-st programs with st, aic7xxx, sd_mod and
scsi_mod modules on  a redhat 7.1 install w/ all redhat updates ( 2.4.2-2
kernel ), DELL poweredge 300 with Sony SDT9000 tape drive.

tar (GNU tar) 1.13.19
dump 0.4b21
mt-st v. 0.5b

cat /proc/scsi/aic7xxx/1
Adaptec AIC7xxx driver version: 5.2.4/5.2.0
Compile Options:
  TCQ Enabled By Default : Enabled
  AIC7XXX_PROC_STATS     : Enabled

Adapter Configuration:
           SCSI Adapter: Adaptec AIC-7850 SCSI host adapter
                           Narrow Controller at PCI 0/16/0
    PCI MMAPed I/O Base: 0xfe100000
 Adapter SEEPROM Config: SEEPROM found and used.
      Adaptec SCSI BIOS: Disabled
                    IRQ: 14
                   SCBs: Active 0, Max Active 1,
                         Allocated 31, HW 3, Page 255
             Interrupts: 4757640
      BIOS Control Word: 0x0000
   Adapter Control Word: 0x0055
   Extended Translation: Disabled
Disconnect Enable Flags: 0x00ff
 Tag Queue Enable Flags: 0x0000
Ordered Queue Tag Flags: 0x0000
Default Tag Queue Depth: 32
    Tagged Queue By Device array for aic7xxx host instance 1:
      {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0}
    Actual queue depth per device for aic7xxx host instance 1:
      {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}

Statistics:

(scsi1:0:0:0)
  Device using Narrow/Sync transfers at 10.0 MByte/sec, offset 15
  Transinfo settings: current(25/15/0/0), goal(25/15/0/0), user(25/15/0/0)
  Total transfers 4754750 (139 reads and 4754611 writes)
             < 2K      2K+     4K+     8K+    16K+    32K+    64K+   128K+
   Reads:       0       0       0       0       0     139       0       0
  Writes:       0       0       0 4638228  116383       0       0       0

cat /proc/scsi/scsi
Attached devices:
Host: scsi0 Channel: 00 Id: 00 Lun: 00
  Vendor: QUANTUM  Model: ATLAS V  9 WLS   Rev: 0201
  Type:   Direct-Access                    ANSI SCSI revision: 03
Host: scsi0 Channel: 00 Id: 01 Lun: 00
  Vendor: QUANTUM  Model: ATLAS V  9 WLS   Rev: 0201
  Type:   Direct-Access                    ANSI SCSI revision: 03
Host: scsi1 Channel: 00 Id: 00 Lun: 00
  Vendor: SONY     Model: SDT-9000         Rev: 0400
  Type:   Sequential-Access                ANSI SCSI revision: 02

/var/log/messages:
Jun 12 03:46:47  kernel: scsi : aborting command due to timeout :
pid 0, scsi1, channel 0, id 0, lun 0 Write (6) 01 0
0 00 14 00
Jun 12 03:46:47  kernel: (scsi1:0:0:0) SCSISIGI 0xe6, SEQADDR
0x11b, SSTAT0 0x2, SSTAT1 0x13
Jun 12 03:46:47  kernel: (scsi1:0:0:0) SG_CACHEPTR 0x0, SSTAT2
0x10, STCNT 0x1ede
Jun 12 03:46:49  kernel: SCSI host 1 abort (pid 0) timed out -
resetting
Jun 12 03:46:49  kernel: SCSI bus is being reset for host 1
channel 0.
Jun 12 04:01:49  kernel: SCSI host 1 abort (pid 0) timed out -
resetting
Jun 12 04:01:49  kernel: SCSI bus is being reset for host 1
channel 0.
Jun 12 04:01:52  kernel: SCSI host 1 channel 0 reset (pid 0) timed
out - trying harder
Jun 12 04:01:52  kernel: SCSI bus is being reset for host 1
channel 0.
Jun 12 04:01:52  kernel: (scsi1:0:0:0) Synchronous at 10.0
Mbyte/sec, offset 15.
Jun 12 04:01:52  kernel: st0: Error with sense data: Info
fld=0x14, Current st09:00: sense key Unit Attention
Jun 12 04:01:52  kernel: Additional sense indicates Power
on,reset,or bus device reset occurred






_______________________________________________
Linux-PowerEdge mailing list
Linux-PowerEdge at dell.com
http://lists.us.dell.com/mailman/listinfo/linux-poweredge



More information about the Linux-PowerEdge mailing list