[brlug-general] tape drive issues

[-ray]

Ronnie,

A few things come to mind.  First check the block size.  I remember having 
a problem (probably on AIX though) of a tape drive taking WAY too long, 
turned out to be a block size mismatch.  You may be able to check with mt 
-f /dev/st0 status.  I don't know what it should be though haha...maybe 
check the tape drive docs and experiment.

I'm guessing this drive is ancient... any firmware updates for it? 
Probably not doing encryption either.   I know RHEL3 is 2.4 kernel based, 
but i'd try with a 2.6 kernel (boot ubuntu or something) and see if the 
problem persists.  There are probably significant changes b/t 2.4 and 2.6.

I'm sure you tried the usual tape suspects already.  Run a new cleaning 
cartridge through the drive, try a new tape, change scsi id, try a 
different cable, different terminator, different scsi card.  try all the 
above on a different machine.   The only thing more fun than 
troubleshooting SCSI problems, is troubleshooting SCSI tape problems haha.

ray



On Fri, 12 Oct 2007, Ronnie Gilkey wrote:

> Hey all,
>
> I'm having some issues with a Benchmark DLT1 tape drive.  What's
> happening is that a daily backup of about 20 GB is taking about 9 hours
> to run.  This is way too long for that little bit of data.  I've checked
> system usage throughout that period with sar reports and the system
> maintains a level of about 96% idle, so the system isn't busy.  For most
> tape modern tape drives 20 GB should take about an hour to backup.  I've
> also checked dmesg for errors, there aren't a lot just these:
>
> st0: Block limits 2 - 16777214 bytes.
> st0: Error with sense data: Info fld=0x2800, Current st09:00: sense key
> Not Ready
> Additional sense indicates Logical unit not ready,initializing cmd. required
> st0: Error with sense data: Current st09:00: sense key Not Ready
> Additional sense indicates Logical unit not ready,initializing cmd. required
> st0: Error 250400f4 (sugg. bt 0x20, driver bt 0x5, host bt 0x4).
> st0: Error 250400f4 (sugg. bt 0x20, driver bt 0x5, host bt 0x4).
> st0: Error on write filemark.
> st0: Error 250400f4 (sugg. bt 0x20, driver bt 0x5, host bt 0x4).
>
> Has anyone else had any problems with this type of tape drive?  Any
> suggestions?  I've posted the scsi_info & sdparm output below.
>
> Thanks,
>
> Ronnie
>
> [root at myhost root]# scsi_info /dev/st0
> SCSI_ID="5,0,0"
> MODEL="BNCHMARK DLT1"
> FW_REV="5538"
> [root at myhost src]# sdparm --all /dev/st0
>    /dev/st0: BNCHMARK  DLT1              5538  [tape]
> Read write error recovery mode page:
>  AWRE        0  [cha: n, def:  0]
>  ARRE        0  [cha: n, def:  0]
>  TB          0  [cha: n, def:  0]
>  RC          0  [cha: n, def:  0]
>  EER         1  [cha: n, def:  1]
>  PER         0  [cha: y, def:  0]
>  DTE         0  [cha: n, def:  0]
>  DCR         0  [cha: n, def:  0]
>  RRC        32  [cha: n, def: 32]
>  COR_S       0  [cha: n, def:  0]
>  HOC         0  [cha: n, def:  0]
>  DSOC        0  [cha: n, def:  0]
>  WRC        15  [cha: n, def: 15]
> Control mode page:
>  TST         0  [cha: n, def:  0]
>  TMF_ONLY    0  [cha: n, def:  0]
>  D_SENSE     0  [cha: n, def:  0]
>  GLTSD       0  [cha: n, def:  0]
>  RLEC        0  [cha: y, def:  0]
>  QAM         0  [cha: n, def:  0]
>  QERR        0  [cha: n, def:  0]
>  RAC         0  [cha: n, def:  0]
>  UA_INTLCK   0  [cha: n, def:  0]
>  SWP         0  [cha: n, def:  0]
>  ATO         0  [cha: n, def:  0]
>  TAS         0  [cha: n, def:  0]
>  AUTOLOAD    0  [cha: n, def:  0]
> Data compression (SSC) mode page:
>  DCE         1  [cha: y, def:  1]
>  DCC         1  [cha: n, def:  1]
>  DDE         1  [cha: n, def:  1]
>  RED         0  [cha: n, def:  0]
>  COMPR_A    16  [cha: n, def: 16]
>  DCOMPR_A   16  [cha: y, def: 16]
> Device configuration (SSC) mode page:
>  CAF         0  [cha: n, def:  0]
>  ACT_F       0  [cha: n, def:  0]
>  ACT_P       0  [cha: n, def:  0]
>  WOBFR       0  [cha: n, def:  0]
>  ROBER       0  [cha: n, def:  0]
>  WDT       200  [cha: y, def:200]
>  OBR         0  [cha: n, def:  0]
>  LOIS        1  [cha: n, def:  1]
>  RSMK        0  [cha: n, def:  0]
>  AVC         0  [cha: n, def:  0]
>  SOCF        0  [cha: n, def:  0]
>  ROBO        0  [cha: n, def:  0]
>  REW         0  [cha: n, def:  0]
>  GAP_S       0  [cha: n, def:  0]
>  EOD_D       0  [cha: n, def:  0]
>  EEG         1  [cha: n, def:  1]
>  SEW         1  [cha: y, def:  1]
>  SWP_T       0  [cha: n, def:  0]
>  BAML        0  [cha: n, def:  0]
>  BAM         0  [cha: n, def:  0]
>  OBSAEW      0  [cha: n, def:  0]
>  SDCA        1  [cha: y, def:  1]
>  WTRE        0  [cha: n, def:  0]
>  OIR         0  [cha: n, def:  0]
>  ROR         0  [cha: n, def:  0]
>  ASOCWP      0  [cha: n, def:  0]
>  PERSWP      0  [cha: n, def:  0]
>  PRMWP       0  [cha: n, def:  0]
> Medium partition (SSC) mode page:
>  MAX_AP      0  [cha: n, def:  0]
>  APD         0  [cha: n, def:  0]
>  FDP         0  [cha: n, def:  0]
>  SDP         0  [cha: n, def:  0]
>  IDP         0  [cha: n, def:  0]
>  PSUM        0  [cha: n, def:  0]
>  POFM        0  [cha: n, def:  0]
>  CLEAR       0  [cha: n, def:  0]
>  ADDP        0  [cha: n, def:  0]
>  MFR         1  [cha: n, def:  1]
>  PART_U      0  [cha: n, def:  0]
> Informational exceptions control mode page:
>  PERF        0  [cha: n, def:  0]
>  EBF         0  [cha: n, def:  0]
>  EWASC       0  [cha: n, def:  0]
>  DEXCPT      1  [cha: n, def:  1]
>  TEST        0  [cha: n, def:  0]
>  EBACKERR    0  [cha: n, def:  0]
>  LOGERR      0  [cha: n, def:  0]
>  MRIE        3  [cha: n, def:  3]
>  INTT        0  [cha: n, def:  0]
>  REPC        0  [cha: n, def:  0]
>
>

-- 
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Ray DeJean  				       	 http://www.r-a-y.org
Systems Engineer                    Southeastern Louisiana University
IBM Certified Specialist  	      AIX Administration, AIX Support
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=