TSM backup failure - sqlerrmc 19??

[agronaught]

I have been hunting through pages all over the place and have the backup to the point that connections are initiated, the backup fails and nothing is logged in the tsmdbmgr.log.

I'm stuck and need a little guidance. I'm including all the relevant information here in the hope it tickles somebodies memory...


For reference this is a clean install of v6.

The following messages are logged in the activity log:

09/02/10 15:23:31 ANR2017I Administrator ADMIN issued command: BACKUP DB
TYPE=FULL DEVCLASS=LTOTAPE SCRATCH=YES (SESSION: 6)
09/02/10 15:23:32 ANR4559I Backup DB is in progress. (SESSION: 6)
09/02/10 15:23:32 ANR0984I Process 7 for DATABASE BACKUP started in the
BACKGROUND at 03:23:32 PM. (SESSION: 6, PROCESS: 7)
09/02/10 15:23:32 ANR2280I Full database backup started as process 7.
(SESSION: 6, PROCESS: 7)
09/02/10 15:23:32 ANR0406I Session 14 started for node $$_TSMDBMGR_$$
(DB2/LINUXX8664) (Tcp/Ip localhost(38096)). (SESSION: 14)
09/02/10 15:23:32 ANR0403I Session 14 ended for node $$_TSMDBMGR_$$
(DB2/LINUXX8664). (SESSION: 14)
09/02/10 15:23:32 ANR0406I Session 15 started for node $$_TSMDBMGR_$$
(DB2/LINUXX8664) (Tcp/Ip localhost(38097)). (SESSION: 15)
09/02/10 15:23:32 ANR0403I Session 15 ended for node $$_TSMDBMGR_$$
(DB2/LINUXX8664). (SESSION: 15)
09/02/10 15:23:32 ANR2968E Database backup terminated. DB2 sqlcode: -2033.
DB2 sqlerrmc: 19. (SESSION: 6, PROCESS: 7)
09/02/10 15:23:32 ANR0985I Process 7 for DATABASE BACKUP running in the
BACKGROUND completed with completion state FAILURE at
more... (<ENTER> to continue, 'C' to cancel)

03:23:32 PM. (SESSION: 6, PROCESS: 7)

The error '19' has thrown me as there is no reference to this error number in the header file or from what I can find - on the web:

#define DSM_RS_ABORT_RETRY 15
#define DSM_RS_ABORT_NO_LOG_SPACE 16
#define DSM_RS_ABORT_NO_DB_SPACE 17
#define DSM_RS_ABORT_NO_MEMORY 18

#define DSM_RS_ABORT_FS_NOT_DEFINED 20
#define DSM_RS_ABORT_NODE_ALREADY_DEFED 21
#define DSM_RS_ABORT_NO_DEFAULT_DOMAIN 22

I believe the DSMI variables are set correctly:

export DSMI_CONFIG=/home/tsminst1/tsminst1/tsmdbmgr.opt
export DSMI_DIR=/opt/tivoli/tsm/client/api/bin64
export DSMI_LOG=/home/tsminst1/tsminst1

The file permissions are correct on the tsmdbmgr.log file. The file is created when the error occurs but nothing is written to it.

The tsmdbmgr.opt file contains:

SERVERNAME TSMDBMGR_TSMINST1

The dsm.sys file is as follows:

SErvername server_a
COMMMethod TCPip
TCPPort 1500
TCPServeraddress node.domain.company.COM

servername TSMDBMGR_TSMINST1
COMMMethod tcpip
tcpserveraddr localhost
tcpport 1500
passwordaccess generate
passworddir /home/tsminst1/tsminst1
errorlogname /home/tsminst1/tsminst1/tsmdbmgr.log
nodename $$_TSMDBMGR_$$

and I have set the password to TSMDBGR with 'dsmapipw' as outlined in the documentation. The relevant DB2 values being:

TSM management class (TSM_MGMTCLASS) =
TSM node name (TSM_NODENAME) =
TSM owner (TSM_OWNER) =
TSM password (TSM_PASSWORD) =

The deviceclass exists and has available volumes.

The db2diag log contains a number of not so helpful entries, the first (following the DB backup start) gives the same error code from TSM:

2010-02-09-15.35.41.853636+660 E2436022E361 LEVEL: Error
PID : 18326 TID : 47931086219968PROC : db2vend (db2med - 126 (TSMDB1))
INSTANCE: tsminst1 NODE : 000
FUNCTION: DB2 UDB, database utilities, sqluvint, probe:482
DATA #1 : TSM RC, PD_DB2_TYPE_TSM_RC, 4 bytes
TSM RC=0x00000013=19 -- see TSM API Reference for meaning.

Really needing some pointers here...

[kehannigan]

Hi. Abort code 19 means "filespace already defined," which should never happen since the TSM API checks for the filespace's existence before trying to define it. I recommend contacting IBM support for help on this... they'll need to get some traces from both the API and server to figure out what's going on here.

[agronaught]

Thanks.

I've initialised a new instance on the same server and that instance works as expected. Fortunately I found this problem before going live

J.