cancel
Showing results for 
Search instead for 
Did you mean: 

DB2 Archive Backup Problem

Former Member
0 Kudos
202

Dear All,

We are facing one issue from last few months and SAP also not able to provide some solution of this problem. I am sharing my problem with you if you guys have some thoughts then please share with me:

System: ERP 6.0 with EHP4 (ECC-DIMP activated)

DB2: DB2 9.1 Fix pack 7

OS: AIX 6.1

When we are running archive log backup from DB13, it is not happening but if I am running online+archive backup problem then it is running successfully. When we ran archive backup from db13 then it uses db2tapemgr utility to take the backup every time it shows below error:

DBT2065I Using database partition "NODE0000".

DBT2071I Using the automatic generated tape-label "GED20100413125014".

Scanning history.

Rewinding tape.

Reading tape header.

Reading tape header failed. Reason: "Error detected."

DBT2027N Tape has not been used for storing log files before.

The FORCE option has been specified.

Rewinding tape.

Writing tape header.

Writing tape header failed. Reason: "Disk full."

DBT2008N db2tapemgr failed with errors.

If I am running this db2tapemgr utility from OS as well then also problem remains same.

Device address which we are using is /dev/rmt0.1. Also, there is no file system that is full and even tape is also not full. We have tried with different set of tapes but none of them is successful.

Logretain and Logarchmeth1 parameters are set and pointing to proper directory that is the reason online + archive backup is successful.

Also, device permissions are correct. And we are not using any third party backup solution.

Please suggest.

Thanks

Sunny

View Entire Topic
malte_schuenemann
Product and Topic Expert
Product and Topic Expert
0 Kudos

There is a couple of possible reasons - andof locations where to search for additional information.

Disk full conditions may be reported in case of

- disk, or filesystem, has indeed become full - in that case, I would expect a related entry in the OS logs (errpr -a)

- a user limit has been hit - have you checked the db2diag.log, or notify log ?

Also: Which user runs db2tapemgr ? Check user limits of that user, and those of the DB2 administrator.

Have you tried writing some file to tape using native OS tools (dd, tar, cpio, ...) ?

Do you have a 'ls -l' output of the device file entry ?

Malte

Former Member
0 Kudos

Hi,

> Disk full conditions may be reported in case of

> - disk, or filesystem, has indeed become full - in that case, I would expect a related entry in the OS logs (errpr -a)

As I said earlier all file system has sufficient space more than 5 GB.

> - a user limit has been hit - have you checked the db2diag.log, or notify log ?

>

ouput of ulimit -a

time(seconds) unlimited

file(blocks) unlimited

data(kbytes) unlimited

stack(kbytes) unlimited

memory(kbytes) unlimited

coredump(blocks) unlimited

nofiles(descriptors) 2000

threads(per process) unlimited

processes(per user) unlimited

> Also: Which user runs db2tapemgr ? Check user limits of that user, and those of the DB2 administrator.

>

db2tapemgr runs from db2<sid> user. ulimits are un-limited as mentioned above.

> Have you tried writing some file to tape using native OS tools (dd, tar, cpio, ...) ?

tar backup is working fine on tape.

> Do you have a 'ls -l' output of the device file entry ?

bash-3.2# ls -l /dev/rmt0.1

crw-rw-rwT 1 root system 49, 0 May 13 14:40 /dev/rmt0.1

db2diag.log for this backup error:

2010-05-17-14.32.54.378198+330 I200218942A312 LEVEL: Warning

PID : 381306 TID : 1 PROC : db2logmgr (GED) 0

INSTANCE: db2ged NODE : 000

FUNCTION: DB2 UDB, data protection services, sqlpgArchiveLogFile, probe:3108

MESSAGE : Started archive for log file S0005084.LOG.

2010-05-17-14.32.55.332510+330 I200219255A422 LEVEL: Warning

PID : 381306 TID : 1 PROC : db2logmgr (GED) 0

INSTANCE: db2ged NODE : 000

FUNCTION: DB2 UDB, data protection services, sqlpgArchiveLogFile, probe:3170

MESSAGE : Completed archive for log file S0005084.LOG to

/db2/GED/log_archive/db2ged/GED/NODE0000/C0000001/ from

/db2/GED/log_dir/NODE0000/.

2010-05-17-15.55.49.401633+330 E200219678A353 LEVEL: Error

PID : 504278 TID : 1 PROC : db2tapemgr

INSTANCE: db2ged NODE : 000

FUNCTION: DB2 UDB, User Application, User Function, probe:1

DATA #1 : String, 10 bytes

db2tapemgr

DATA #2 : String, 54 bytes

rc=4 (ca:rc=NOT A LOGFILE TAPE,reason=Error detected.)

Please suggest.

Thanks

Sunny

malte_schuenemann
Product and Topic Expert
Product and Topic Expert
0 Kudos

I suspect this is an issue to be followed up by SAP Support. Please open a SAP ticket.

Malte

Former Member
0 Kudos

Already opened a message with SAP around two and half month ago but still no solution.

Thanks

Sunny