Re: TSM - CPF414D

I am thinking the contention was that TSM does this backup storage pool 
using two processes.  We have two drives in our 3582-L23 so it uses both 
at the same time.  It actually tried to chunk out both tapes and go on to 
the next about the same time.

09/26/05 00:53:37     ANR8341I End-of-volume reached for LTO volume 03WK1. 
 
                       (SESSION: 11127, PROCESS: 152)  
09/26/05 00:53:49     ANR0515I Process 152 closed volume 03WK1. (SESSION: 
11127,
                       PROCESS: 152)  
09/26/05 00:53:50     ANR8336I Verifying label of LTO volume 03WK1 in 
drive 
                       TAPMLB01 (DRVRSRC). (SESSION: 11127, PROCESS: 152)  
 
09/26/05 00:54:30     ANR8468I LTO volume 03WK1 dismounted from drive 
TAPMLB01 
                       (DRVRSRC) in library TAPMLB01. (SESSION: 11127, 
PROCESS: 
                       152)  
09/26/05 00:54:31     ANR8373I 074: Waiting for the completion of a system 
MLB 
                       operation in device TAPMLB01. (SESSION: 11127, 
PROCESS: 
                       152)  
09/26/05 00:54:42     ANR0515I Process 151 closed volume 04WK1. (SESSION: 
11127,
                       PROCESS: 151)  
09/26/05 00:54:43     ANR1181E astxn.c589: Data storage transaction 
0:2779524794
                       was aborted. (SESSION: 11127, PROCESS: 151)  
09/26/05 00:54:44     ANR2183W dfbackup.c1497: Transaction 0:2779524794 
was 
                       aborted. (SESSION: 11127, PROCESS: 151)  
09/26/05 00:54:44     ANR1280E BACKUP STGPOOL: Process 151 terminated -   
                       transaction aborted. (SESSION: 11127, PROCESS: 151) 
 
09/26/05 00:54:44     ANR0986I Process 151 for BACKUP STORAGE POOL running 
in 
                       the BACKGROUND processed 1859187 items for a total 
of 
                       368,999,694,336 bytes with a completion state of 
FAILURE 
                       at 00:54:44. (SESSION: 11127, PROCESS: 151)  
09/26/05 00:54:48     ANR8468I LTO volume 04WK1 dismounted from drive 
TAPMLB01 
                       (DRVRSRC) in library TAPMLB01. (SESSION: 11127, 
PROCESS: 
                       151)  
09/26/05 00:55:31     ANR8373I 074: Waiting for the completion of a system 
MLB 
                       operation in device TAPMLB01.  
09/26/05 00:56:31     ANR8373I 074: Waiting for the completion of a system 
MLB 
                       operation in device TAPMLB01.  
09/26/05 00:56:55     ANR8328I 074: LTO volume 05WK1 mounted in drive 
TAPMLB01 
                       (DRVRSRC). (SESSION: 11127, PROCESS: 152)  
09/26/05 00:56:55     ANR1340I Scratch volume 05WK1 is now defined in 
storage 
                       pool LTO_3582_WEEK1. (SESSION: 11127, PROCESS: 152) 
 
09/26/05 00:56:55     ANR0513I Process 152 opened output volume 05WK1. 
(SESSION:
                       11127, PROCESS: 152)


And TSM uses the PASE api's to talk to the tape drive.  The PASE tape 
api's SUCK!  They are unreliable.  They are a performance dog.  We only 
get a third of the transfer rate using them that we do with native save 
commands.  IBM admits to the performance issue.  Their solution:  dump 
i5/os and run TSM on other platforms.

Rob Berendt