cancel
Showing results for 
Search instead for 
Did you mean: 

Work process crash after upgrade to Oracle 11g

Former Member
0 Kudos

Hi guys,

I have recently upgraded my Oracle database to Oracle 11.2.0.2. My environment is SAP R/3 4.7 ext 2.00, AIX 5.3 TL 9.

I also have upgrade my SAP kernel to 6.40_EX2 kernel (the lates one).

I have tested some of advanced compression features including index compression and table compression. But couple times (when this system is idle), suddenly SAP work process is crashed without notice.

Is there any experience about this ? Any suggestion would be appreciated.

Thannk you

Accepted Solutions (1)

Accepted Solutions (1)

Former Member
0 Kudos

Please check the corresponding workprocess trace dev_wX, dev_disp and SM21. Also check the alert log and the user traces in user_dump_dest.

There must be a message somewhere.

Kind regards, Michael

Answers (2)

Answers (2)

former_member184473
Active Contributor
0 Kudos

Hello,

Are you using at least December SBP? As per note [1436352|http://service.sap.com/sap/support/notes/1436352]:

o Oracle Database Release 11.2.0.2 or higher +

SAP December 2010 Bundle Patch or later

(to achieve good performance against OLTP compressed tables)

Regards,

Eduardo Rezende

Former Member
0 Kudos

hi all,

i am still analyzing dev_x log files right now (just as your suggest)

ardhian

Former Member
0 Kudos

Post one here..

Mark

Former Member
0 Kudos

Hi All,

This is the log file for dev_w0

================

M Sat Mar 26 00:36:16 2011

M in_ThErrHandle: 1

M ThIErrHandle: new stat of W0 is WP_SHUTDOWN

M ThIErrHandle: I'm during shutdown

M PfStatDisconnect: disconnect statistics

M Entering ThSetStatError

B db_sqlbreak() = 1

M ThIErrHandle: don't try rollback again

M ThShutDownServer: shutdown server

M ThExecShutDown: perform exclusive shutdown actions

M ThCheckComOrRb (event=1, full_commit=1)

M ThCallHooks: call hook >ASTAT-collect commit handling< for event BEFORE_COMMIT

M ThCallHooks: call hook >rsts_before_commit< for event BEFORE_COMMIT

M SosSearchAnchor: search anchor for 10

M SosSearchAnchor: invalid tid/mode T-1/M255

M ThCheckComOrRb (event=3, full_commit=1)

M ThCallHooks: call hook >ThVBICmRbHook< for event AFTER_COMMIT

M ThVBICmRbHook: called for commit

M ThCallHooks: call hook >ThNoClearPrevErr< for event AFTER_COMMIT

M ThNoClearPrevErr: clear prev no err

M ThCallHooks: call hook >rsts_after_commit< for event AFTER_COMMIT

M SosSearchAnchor: search anchor for 10

M SosSearchAnchor: invalid tid/mode T-1/M255

M ThCallHooks: call hook >dyKeyTableRest< for event AFTER_COMMIT

M ThCallHooks: call hook >SpoolHandleHook< for event AFTER_COMMIT

M SosSearchAnchor: search anchor for 2

M SosSearchAnchor: invalid tid/mode T-1/M255

M ThUsrDelEntry (*, *, emaprd2_WPR_10 ) o.k.

M ThICommit3: full commit, set time, keep resources, redispatch

M ThICommit3: commit and keep resources

M ThCheckComOrRb (event=1, full_commit=0)

M ThCallHooks: call hook >ASTAT-collect commit handling< for event BEFORE_COMMIT

M ThCallHooks: call hook >rsts_before_commit< for event BEFORE_COMMIT

M SosSearchAnchor: search anchor for 10

M SosSearchAnchor: invalid tid/mode T-1/M255

M ThCheckComOrRb (event=3, full_commit=0)

M ThCallHooks: call hook >ThVBICmRbHook< for event AFTER_COMMIT

M ThVBICmRbHook: called for commit

M ThCallHooks: call hook >ThNoClearPrevErr< for event AFTER_COMMIT

M ThNoClearPrevErr: clear prev no err

M ThCallHooks: call hook >rsts_after_commit< for event AFTER_COMMIT

M SosSearchAnchor: search anchor for 10

M SosSearchAnchor: invalid tid/mode T-1/M255

M ThCallHooks: call hook >dyKeyTableRest< for event AFTER_COMMIT

M ThCallHooks: call hook >SpoolHandleHook< for event AFTER_COMMIT

M SosSearchAnchor: search anchor for 2

M SosSearchAnchor: invalid tid/mode T-1/M255

M ThAlarm: set alarm to 600 sec

M ThICommit3 o.k.

M ThExecShutDown: ThUsrDelEntry o.k.

M ThExecShutDown: called rsau_log_system_stop

M PfStatIndInit: Initializing Index-Record

M PfWriteIntoFile: copied shared buf (0 bytes) to local buf

M PfTimeFactor: new value: 25200

M Entering ThReadDetachMode

M call ThrShutDown (1)...

B Disconnecting from ALL connections:

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

B 000 000 R/3 000000000 INACTIVE NO NO YES NO NO 000 255 255 20110326 002910 emaprd2

C Disconnecting from connection 0 ...

C Close user session (con_hdl=0,svchp=113176338,usrhp=1131b6f80)

C Detaching from DB Server (con_hdl=0,svchp=113176338,srvhp=1131773e0)

C Now connection 0 is disconnected

B Disconnected from connection 0

B statistics db_con_commit (com_total=107, com_tx=58)

B statistics db_con_rollback (roll_total=1, roll_tx=0)

M ***LOG Q02=> wp_halt, WPStop (Workproc 0 336276) [dpuxtool.c 318]

================

Any suggestion ?

ardhian

http://sapbasis.wordpress.com

http://ardhian.wordpress.com

Former Member
0 Kudos

Well there is no error here, the work process is shutdown, but we cannot see why.

Please check the corresponding workprocess trace dev_wX, dev_disp and SM21. Also check the alert log and the user traces in user_dump_dest.

I assume there are no relevant lines just before the output shown here? Well then i would like to see the corresponding dev_disp lines and the SM21 entries.

Cheers Michael

Former Member
0 Kudos

hi guys,

The problem is solved right now. I have just download the latest one for dw (dispwork) files from SAP (just recently released per March 2011). I think there is huge bug on dw (dispwork) files (released on March 5 2011).

It stable until now.

Thannk you.

ardhian

http://sapbasis.wordpress.com

http://ardhian.wordpress.com

Former Member
0 Kudos

Hi,

what's in your dev_w* trace files when the processes crash..

Mark