WP status running without recognizable actions

[Former Member]

Hi gurus!

We detect a very strange behaviour in our production system.

Suddenly, the workprocesses change to running status but without any information about the execution in sm50, no report, no user, no sql statement...

The WPs cannot be restarted and the only way to solve it (at least that we found) is to restart the instance.

Our system is SAP R/3 Enterprise Basis 6.20 with DB v8 database.

Today is the second day that we bear this issue.

Can anybody to advise about a solution?

Thank you very much in advanced,

Edorta Ramos.

[Former Member]

Thank you for your replies,

Hari, the db we have installed is not oracle but DB2.

I have already checked the note 801844 but this is not our case, the ora-1224 appears when we start shutting down the system.

thank you very much in advanced,

Edorta.

[Former Member]

Hi

Import the new database interface (dbdb2slib) Patch level 1774 or higher. See Note 801844

Cheers

Shaji

[Former Member]

Hi Edorta,

Check Dev_wp<x> log and also try activating the trance and check the trace file for that workprocess.

Hope this helps you to identify problem else post these files.

Regards,

Vamshi.

[Former Member]

Thanks for your reply vamshi but I have already checked the traces for the WPs but I am not able to find nothing relevant.

This is one of the traces:

trc file: "dev_w2", trc level: 1, release: "640"

***

• ACTIVE TRACE LEVEL 1*

• ACTIVE TRACE COMPONENTS all, M*

***

M sysno 00

M sid SP1

M systemid 324 (IBM RS/6000 with AIX)

M relno 6400

M patchlevel 0

M patchno 175

M intno 20020600

M make: single threaded, ASCII, 64 bit

M pid 1724878

M

M

M Mon Mar 17 09:29:37 2008

M calling db_connect ...

C

C Mon Mar 17 09:29:39 2008

C Callback functions for dynamic profile parameter registered

C dbs/db2/use_drda = 1 -> DRDA will be used

C New Functions of DB2 V8 are switched on

C dbs/db2/use_accounting = 1 -> DB2 accounting is switched on

C dbs/db2/opt2_hint = 0 -> implicit 'optimize for 1 rows' is switched on

C dbs/db2/rs_by_hint = 1 -> isolation level RS by hint is switched on

C

C Mon Mar 17 09:29:40 2008

C SQL DRIVER VERSION is "08.02.0005"

C DB2 Connect Version less than 09.00.0000: Use FAT client.

C DB2Connect driver identified as FAT CLIENT

C DBSLHA: Got Failover profile /usr/sap/SP1/SYS/global/connect.ini

C DBSLHA: Using new Failover Support

C DBSLHA: Using user( g301349) and password(<pwd>) from password service.

C DBSLHA: Using user( g301349) and password(<pwd>) from password service.

C DBSLHA: Using user( g301349) and password(<pwd>) from password service.

C Connecting to <SP1_on_ibmdbp_sysplex> on connection 0 ...

C

C Mon Mar 17 09:29:41 2008

C DB2TRC: 000000 CLI_ALLOC_ENV 1

C DB2TRC: 000000 CLI_ALLOC_DBC 1

C

C Mon Mar 17 10:14:18 2008

C DB2TRC: 000000 CLI_CONNECT 0

C Now I'm connected to DB2 (8.1.5)

C COLLECTION ID used is "SAP08.02.0005"

C SQL DRIVER NAME is "libdb2.a"

C SQL DBMS NAME is "DB2"

C SQL DBMS VERSION is "08.01.0005"

C DATABASE NAME(DB2 Connect DCS database name) is "SP1 "

C Profile parameter SWITCH dbs/db2/cli_trace=0

C EBCDIC CCSID retrieved from Monitor table DB2QWPT2 is 500 .

C Your DB2 system works with ASCII code page 819 and EBCDIC code page 500

C [dbdb2dic.c 1431] INFO Profile: SDB2_DEBUG=<UNSET>

C

C [dbdb2dic.c 1455] INFO Envrmnt: sdb2_debug=<UNSET>

C

C [dbdb2dic.c 1455] INFO Envrmnt: SDB2_DEBUG=<UNSET>

C

C DATABASE LOCATION NAME is "SAPDB_P "

B Connection 0 opened (DBSL handle 0)

B Wp Hdl ConName ConId ConState TX PRM RCT TIM MAX OPT Date Time DBHost

B 000 000 R/3 000000000 ACTIVE NO YES NO 000 255 255 20080317 092939 SP1_on_ibmdbp_sy

M db_connect o.k.

*M ICT: exclude compression: .zip,.cs,.rar,.arj,.z,.gz,.tar,.lzh,.cab,.hqx,.ace,.jar,.ear,.war,.css,.pdf,.js,.gzip,.uue

I

I Mon Mar 17 10:14:19 2008

I MtxInit: 2 0 0

M

M Mon Mar 17 10:14:20 2008

M SHM_PRES_BUF###(addr: 70000000e2f0000, size: 39000000)

M SHM_ROLL_AREA##(addr: 700000031e47000, size: 131072000)

M SHM_PAGING_AREA##(addr: 700000039b48000, size: 134217728)

M SHM_ROLL_ADM###(addr: 700000010823000, size: 2672386)

M

M Mon Mar 17 10:14:32 2008

M SHM_PAGING_ADM##(addr: 700002120000000, size: 656416)

M ThCreateNoBuffer##allocated 324144 bytes for 1000 entries at 700000010ab1000

M ThCreateNoBuffer##index size: 3000 elems

M ThCreateVBAdm##allocated 7040 bytes (50 server) at 700000010b02000

X EmInit: MmSetImplementation( 2 ).

X EM/TOTAL_SIZE_MB = 262144

B db_con_shm_ini: WP_ID = 2, WP_CNT = 15, CON_ID = 2000035

B

B Mon Mar 17 10:14:39 2008

B dbtbxbuf: Buffer TABL (addr: 700002130000100, size: 10985200, end: 700002130a79ff0)

B dbtbxbuf: Buffer TABLP (addr: 700000016910100, size: 30720000, end: 70000001865c100)

B dbexpbuf: Buffer EIBUF (addr: 700000018662108, size: 40960000, end: 70000001ad72108)

B dbexpbuf: Buffer ESM (addr: 700002140000108, size: 4194304, end: 700002140400108)

B dbexpbuf: Buffer CUA (addr: 70000001ad74108, size: 8192000, end: 70000001b544108)

B dbexpbuf: Buffer OTR (addr: 700002150000108, size: 4194304, end: 700002150400108)

M rdisp/reinitialize_code_page -> 0

M icm/accept_remote_trace_level -> 0

M rdisp/no_hooks_for_sqlbreak -> 0

M CCMS: AlInitGlobals : alert/use_sema_lock = TRUE.

C

C Mon Mar 17 10:14:40 2008

*C *** ERROR => DB2 Call 'CLI_EXECUTE' Error: sqlcode = -1224 : [IBM][CLI Driver] SQL1224N A database agent could not be started to servi*

The last line corresponds with the system shutdown.

Any suggestions?

Thank you very much in advanced,

Edorta.

[Former Member]

Hi Edorta,

Most of the times when I have seen such a problem it is an archiver stuck. Can you refer to your ORACLE Alert file and check for write failure in the archive processes?

Check the filesystem for ORACLE Archive logs. If this happens, all changeing transactions will be hanging until this archiver stuck is fixed.

Regards,

Hari.