Dialog WP went to New state

Hi all,

All the dialog WP went into new state in one of the application server. I tried to restart it through dpmon but it still went to new state. I have checked the R3trans-d it also works fine. Kindly suggest on it.

Regards,

Karthik.R

SAP Managed Tags:
SAP NetWeaver

You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.

Comment

Accepted Solutions (0)

Answers (2)

Hi Karthik,

better kill all the work process forcefully then it will automatically works. I had the same issue I've killed forcefully it works .

hope for you also it works.

Thanks,

You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.

Comment

Kindly check db log file once. might be it is filled.

Hello Karthik,

could you please share a screenshot (or dpmon output) about SM50 and the trace of one affected work process?

Regards,
Alwina

You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.

Comment

Hi Alwina,

Please find the screenshot of dpmon and the log file below.

ABAP Program SAPMHTTP .

A Source SAPMHTTP Line 19.

A Error Code TIME_OUT.

A Module $Id: //bas/721_REL/src/krn/th/thhiserv.c#6 $ SAP.

A Function ThTimeout Line 1848.

A ** RABAX: level LEV_RX_STDERR completed.

A ** RABAX: level LEV_RX_RFC_ERROR entered.

A ** RABAX: level LEV_RX_RFC_ERROR completed.

A ** RABAX: level LEV_RX_RFC_CLOSE entered.

A ** RABAX: level LEV_RX_RFC_CLOSE completed.

A ** RABAX: level LEV_RX_IMC_ERROR entered.

A ** RABAX: level LEV_RX_IMC_ERROR completed.

A ** RABAX: level LEV_RX_DATASET_CLOSE entered.

A ** RABAX: level LEV_RX_DATASET_CLOSE completed.

A ** RABAX: level LEV_RX_ERROR_SAVE entered.

A ** RABAX: level LEV_RX_ERROR_SAVE completed.

A ** RABAX: level LEV_RX_ERROR_TPDA entered.

A ** RABAX: level LEV_RX_ERROR_TPDA completed.

A ** RABAX: level LEV_RX_PXA_RELEASE_RUDI entered.

A ** RABAX: level LEV_RX_PXA_RELEASE_RUDI completed.

A ** RABAX: level LEV_RX_LIVE_CACHE_CLEANUP entered.

M RmCleanUpResources3: hdr/tbl/ext_tbl/free/keep = 7fc80d20f940/7fbe4aa8a358/0/0/1

A ** RABAX: level LEV_RX_LIVE_CACHE_CLEANUP completed.

A ** RABAX: level LEV_RX_END entered.

A ** RABAX: level LEV_RX_END completed.

M SigIRegisterHook: cleared hook for signal SIGSEGV

M SigIRegisterHook: cleared hook for signal SIGBUS

A ** RABAX: end RX_RETURN

A Time limit exceeded..

M ThTimeout: no online user, delete mode

M DpVmcGetVmByTmAdm: no VM found for T66/M0/I0 (locked VMs ignored)

M ThIAMDel: delete tid/mode 66/0 (th_errno 52, release 1)

M ThIDeleteMode (66, 0, 3, ><, 0, 255, TRUE)

M ThIDeleteMode: no modes found ..

M no sub modes

M ThModeChainCheck: mode chain of T66 o.k.

M ThCheckMemoryState (0, 0, 1)

M ThIAMDel2: entering ThSetStatError

M ThIGetT100: read tskh msg 52 from msg area 14 (db rel 731)

M ThPlgCleanAll: clean all plugins of T66/M0/I0 (th_errno=55, level=2)

M ThPlgClean: clean plugin_hdl=0 of T66/M0 (th_errno=55)

M ThPlgClean: call cleanup handler

M ThPlgClean: clean server plugin

M DpSetContextCounter: num_used_contexts[1] = 5

M ThPlgClean: errno=55, state=DP_PLUGIN_CANCELED, cancel plugins

M ThPlgClean: cancel read MPI<f05b>12

M ThPlgClean: cancel write MPI<f05c>15

M ThPlgSendErrTxt2: plugin already free or canceled

M ThIAMDel2: last active mode (0) delete tid (66)

M ThIUsrDel (52, 1)

M ThCheckMemoryState (0, 0, 1)

M ThModeChainCheck: mode chain of T66 o.k.

M ThIGetT100: read tskh msg 52 from msg area 14 (db rel 731)

M ThEppGetConnectionCounter: read connectionCounter 1 from epp 0

I *** ERROR => invalid MPI handle: 0 [mpixx.c 3372]

I {root-id=0017A477180C1EE49DC7E4A607E696A2}_{conn-id=0017A477180C1EE49DC7E4A605D0F6A2}_1

M ThPlgCleanAll: clean all plugins of T66/M0/I0 (th_errno=55, level=2)

M ThPlgSendErrTxt2: plugin already free or canceled

M ThResFree: free resources of U885 M0 I0 (normal mode, complete free) at level 3, errno=52, db_action=TH_DB_ROLLBACK

M ThResFree: reset spa state for user T66/U885/M0

M ThCheckMemoryState (0, 0, 1)

M ThCallHooks: call hook >abTpdaThHook_EModeEnded< for event BEFORE_SESSION_CANCEL

M ThCallHooks: hook >abTpdaThHook_EModeEnded< o.k.

M ThCallHooks: call hook >ITSP: EndOfSession< for event BEFORE_SESSION_CANCEL

M ThCallHooks: hook >ITSP: EndOfSession< o.k.

M ThCallHooks: call hook >abRtmClearSession< for event BEFORE_SESSION_CANCEL

M ThCallHooks: hook >abRtmClearSession< o.k.

M ThCallHooks: call hook >HttpClearSession< for event BEFORE_SESSION_CANCEL

M ThCallHooks: hook >HttpClearSession< o.k.

M ThCallHooks: call hook >SpoolHandleHook< for event BEFORE_SESSION_CANCEL

M ThCallHooks: hook >SpoolHandleHook< o.k.

M ThIRollBack: call db_rollback

M ThCheckComOrRb (event=0, full_commit=1)

M ThCallHooks: call hook >ab_SosRfcBeforeRollback< for event BEFORE_ROLLBACK

M ThCallHooks: hook >ab_SosRfcBeforeRollback< o.k.

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

M PfAStatRevertCollected: nothing to do

M ThCallHooks: hook >ASTAT-collect commit handling< o.k.

M ThCallHooks: call hook >dyKeyTableReset< for event BEFORE_ROLLBACK

M ThCallHooks: hook >dyKeyTableReset< o.k.

M ThCallHooks: call hook >SpoolHandleHook< for event BEFORE_ROLLBACK

M ThCallHooks: hook >SpoolHandleHook< o.k.

M ThCheckComOrRb (event=2, full_commit=1)

M ThCallHooks: call hook >ThVBICmRbHook< for event AFTER_ROLLBACK

M ThVBICmRbHook: called for rollback

M ThVBIRollBackRedo: got 0 entries to delete again

M ThCallHooks: hook >ThVBICmRbHook< o.k.

M ThCallHooks: call hook >ThNoClearPrevErr< for event AFTER_ROLLBACK

M ThNoClearPrevErr: clear prev no err

M ThCallHooks: hook >ThNoClearPrevErr< o.k.

M ThCallHooks: call hook >ThNoResetMutex< for event AFTER_ROLLBACK

M ThNoCheckMutex: check for remaining locked intervals

M ThCallHooks: hook >ThNoResetMutex< o.k.

M ThCallHooks: call hook >rstshwr_hook_was_rollback< for event AFTER_ROLLBACK

M ThCallHooks: hook >rstshwr_hook_was_rollback< o.k.

M ThCallHooks: call hook >SpoolHandleHook< for event AFTER_ROLLBACK

M ThCallHooks: hook >SpoolHandleHook< o.k.

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

M ThICommit3: commit and free resources

M ThCheckComOrRb (event=1, full_commit=1)

M ThCallHooks: call hook >ab_SosShmBeforeCommit< for event BEFORE_COMMIT

M ThCallHooks: hook >ab_SosShmBeforeCommit< o.k.

M ThCallHooks: call hook >ab_SosRfcBeforeCommit< for event BEFORE_COMMIT

M ThCallHooks: hook >ab_SosRfcBeforeCommit< o.k.

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

M ThCallHooks: hook >ASTAT-collect commit handling< o.k.

M ThCallHooks: call hook >rstshpc_hook_prepare_for_commit< for event BEFORE_COMMIT

M ThCallHooks: hook >rstshpc_hook_prepare_for_commit< o.k.

M ThCheckComOrRb (event=3, full_commit=1)

M ThCallHooks: call hook >ThVBICmRbHook< for event AFTER_COMMIT

M ThVBICmRbHook: called for commit

M ThCallHooks: hook >ThVBICmRbHook< o.k.

M ThCallHooks: call hook >ThNoClearPrevErr< for event AFTER_COMMIT

M ThNoClearPrevErr: clear prev no err

M ThCallHooks: hook >ThNoClearPrevErr< o.k.

M ThCallHooks: call hook >dyKeyTableRest< for event AFTER_COMMIT

M ThCallHooks: hook >dyKeyTableRest< o.k.

M ThCallHooks: call hook >rstshwc_hook_was_commit< for event AFTER_COMMIT

M ThCallHooks: hook >rstshwc_hook_was_commit< o.k.

M ThCallHooks: call hook >SpoolHandleHook< for event AFTER_COMMIT

M ThCallHooks: hook >SpoolHandleHook< o.k.

M ThrClearMemory o.k.

M ThCallHooks: call hook >db_x_short_mem_sync_point< for event AFTER_STORAGE_FREE

M ThCallHooks: hook >db_x_short_mem_sync_point< o.k.

M ThCallHooks: call hook >rstshwsf_hook_was_storage_free< for event AFTER_STORAGE_FREE

M ThCallHooks: hook >rstshwsf_hook_was_storage_free< o.k.

M ThAlarm: set alarm to 1200 sec

M ThICommit3 o.k.

M PfStatDisconnect: disconnect statistics

M ThDealComm: T66/U885/M0 has no cpic conns

M clean up all internal modes of T66/U885/M0

M call ab_endmode (END_MODE_CLEAR_ALL)

M RstrNotifyUserChange: user/client = (SAPSYS /000)

M ***LOG R47=> ThResFree, delete (001052) [thxxmode.c 1344]

M ThGetEnqInfo: no enq info found

M ThRmLocQueue: no async request found

M ThPlgCleanAll: clean all plugins of T66/M0/I0 (th_errno=52, level=3)

M ThPageFree: free (885,0,0, mode = 3)

M ThRollFree: free (885,0,0, mode = 3)

M ThEppResetAdm: invalidate epp adm for T66/M0 at 7fbe4aa89db8

M ThRmFree: cleanup resources for T66 M0

M RmCleanUpResources3: hdr/tbl/ext_tbl/free/keep = 7fc80d20f940/7fbe4aa8a358/0/1/0

M RmCleanUpResources3: call resource handler 8f81cc for key/entry 10/0

M RmInitResourceTblEntry: free resource description at 7fbe4aa8ad08 (ext_tbl=0)

M RmIFreeTable: free resource table at 7fbe4aa8a340 (ext_tbl=0)

M ThRmFree: reset eg state of T66 M0

M ThRmFree cleanup resources for T66

M RmCleanUpResources3: hdr/tbl/ext_tbl/free/keep = 7fc80d215f68/(nil)/1/1/0

M RmCleanUpResources3: no resources registered

M ThRmFree: reset eg state of T66

M ThInvalidateEnqMem: reset enq memory 7fbe4aa88208 of T66/M0

M ThIEmFree: delete hyper context (hyperHdl=7fbe4c468b28)

M ThIEmFree: call EmContextDelete (em_hdl=72)

M ThResFree: save spa pointer 7fc7ff14c680 and install dummy spa

M ThClearMode: clear tid/mode 66/0 (next/prev 0/0)

M ThIUsrDel: set state of term T66 to TM_NEND

M ThISend: (T66/M0, user=SAPSYS , type=0x41, info=0x0, mode_deleted=1)

M type=TASK_SWITCH,DISP

M ThCheckAbapDebugging: no valid user or ABAP not no stack, skip check

M ThJCheckJavaDebugging: return 0

M ThrtGetStrategy: abap strategy ROLL / O.K.

M ThISend: set TH_LOGOFF for T66 (user=SAPSYS ) in state TM_NEND

M ThISend: reset act fields T66/M0 as soon as possible

M ThRqOutCheck: check output for T66/M0

M ThRqOutCheck: o.k.

M ThISend: allowed rq_type of T66/M0 = TH_ALL_RQ

M ThNewWpStat (type=0x10000041, task_switch=1, inline_hold=0, hand_shake=0, debug=0, ..)

M ThNewWpStat: new MODE_REC = MODE_REC_SYS

M ThNewWpStat: new state of T66/M0 = MODE_WAIT,MODE_REC_SYS

M ThNewWpStat: new wp stat: WP_WAIT

rdisp/wp_auto_restart = 0 sec

M rdisp/wp_abap_restart = 0

M ThSetThToolParam : restartable_wp[0] = FALSE

M ThSetThToolParam : restartable_wp[1] = TRUE

M ThSetThToolParam : restartable_wp[2] = TRUE

M ThSetThToolParam : restartable_wp[3] = FALSE

M ThSetThToolParam : restartable_wp[4] = TRUE

M ThSetThToolParam : restartable_wp[5] = TRUE

M ThSetThToolParam : restartable_wp[6] = TRUE

M ThMkReq: detach wp_ca_blk 84

M ThDtWpBlk: detach WP ca_blk 84

M ThISend: complete stat record

M ThCallHooks: call hook >TrThHookFunc< for event COMPLETE_STAT_RECORD

M TrThHookFunc: called for complete stat rec

M ThGetTransactionId: no transaction id found

M TrIGetTransactionId: no transaction-ID received (3)

M ThCallHooks: hook >TrThHookFunc< o.k.

M ThISend: stat_state = TH_STAT_WRITE

M ThISend: th_wp_waits_for_rq = TH_WP_WAITS_FOR_RQ

M make DISP owner of wp_ca_blk 84

M DpRqIPutIntoQ: put elem in queue NOWP, elems=4

M DpRqPutIntoQueue: put request into queue (reqtype 0, prio LOW, rq_id 2302)

M -OUT- sender_id WORK_PROCESS tid 66 wp_ca_blk 84 wp_id 32

M -OUT- action SEND_TO_DP uid 885 appc_ca_blk -1 type NOWP

M -OUT- new_stat WP_WAIT mode 0 len 8 rq_id 2302

M -OUT- req_info LOGOFF

M ThInvalidateActFields: invalidate act fields from T66/M0/U885/no VM

M ThCheckBlks: th_step = TH_WORK_NOSYNC

M ThCheckBlks: release appc_ca_blk 2108

M REL APPC ca_blk 2108

M ThDtAppcBlk: detach APPC ca_blk 2108

M PfClearAllRec: clear statistic area

M PfStatDisconnect: disconnect statistics

M ThCallHooks: call hook >TrThHookFunc< for event BEFORE_TASK_SWITCH

M TrThHookFunc: called for task switch

M ThCallHooks: hook >TrThHookFunc< o.k.

M ThCallHooks: call hook >ThPCleanZombiesCallback< for event BEFORE_TASK_SWITCH

M ThCallHooks: hook >ThPCleanZombiesCallback< o.k.

M ThCallHooks: call hook >ImcResetPtr< for event BEFORE_TASK_SWITCH

M ThCallHooks: hook >ImcResetPtr< o.k.

M ThCallHooks: call hook >HttpResetPtr< for event BEFORE_TASK_SWITCH

M ThCallHooks: hook >HttpResetPtr< o.k.

M ThCallHooks: call hook >ThIDbgCallbackBeforeTaskSwitch< for event BEFORE_TASK_SWITCH

M ThCallHooks: hook >ThIDbgCallbackBeforeTaskSwitch< o.k.

M ThISend: jump back with rc TH_WAIT_FOR_REQUEST

M TskhLoop: switch back to normal tracel level 1

A UpdateProfile (new settings): off,krn_impl,(no dyn check),(no logging),(no stat err),(local generation),(no commit)

A UpdateProfile (new exceptions): (tmp_err),(home_err),(local_err)

A UpdateProfile (new version): 0

M SecAudit(rsauinit): WP attached to existing shared memory.

M SecAudit(RsauShmInit): addr of SHM for Audit.. = 7fc8221df000

M SecAudit(RsauShmInit): addr of RSAUSHM........ = 7fc8221e0000

M SecAudit(RsauShmInit): addr of RSAUSLOTINFO... = 7fc8221e0610

M SecAudit(RsauShmInit): addr of RSAUSLOTS...... = 7fc8221e0620

M ThAlarmHandler: first alarm, just set controls

M ThAlarmHandler: (2)

M ThInsideCriticalSection: waiting for semaphore 42

M ThInsideCriticalSection: waiting for semaphore 42 held by process 19763 for 1134s

M ThAlarmHandler: set CONTROL_TIMEOUT/DP_CONTROL_JAVA_EXIT and break sql

M ThAlarmHandler: return from signal handler

M ThAlarmHandler: (3)

M ThInsideCriticalSection: waiting for semaphore 42

M ThInsideCriticalSection: waiting for semaphore 42 held by process 19763 for 1s

M ThAlarmHandler: 3. ALARM: terminate process (pid=9175, user is T66/M0)

Regards,

Karthik.R

Hi Karthik,

The problem could be with memory pipes so, you should take a restart of your particular instance and clean shared memories.

If it's a non-windows platform, use cleanipc and ipcrm to clean ipcs.

If Windows, a restart might be required.

BTW, have checked file system and DB logs ?

Regards,

Hello Karthik,

there is a problem with the semaphore 42, the work process 25 helds it. Could you please attach dev_w25 of this work process?

Semaphore 42:

42: SEM_STAT_BUF (shared statistic buffer)

as per note "33873 What do the semaphores mean?"

Regards,
Alwina

Hello Divyanshu,

this is the semaphore 42, which is causing the issue, you see it with "-42" for the process, which cannot release it, the trace of this process needs to be checked.

And one do not need to restart the whole instance, only the work process 25 needs to be killed.

Regards,
Alwina

also the work process 45 looks not good, it has the longest runtime and it does not execute any actions (no user, no report)

what is the kernel release? For semaphore 42 there is this KBA:

2007484 - Semaphore 42. Kernel 72x.

yes, may be..

However, I saw this screenshot now and thought it was with all processes.

Thanks for the reply The kernel release is 721_EXT 300 patch

In the attached trace you see:

M ThInsideCriticalSection: waiting for semaphore 42 held by process 19763 for 1s

one can try to kill only WP25 first and see, if it will improve the situation

then you should check the parameter rdisp/sem_recover_level as per KBA "2007484 - Semaphore 42. Kernel 72x.".

Please try to kill WP25 on OS level, if it will not work, then you have to stop the instance and to cleanup the shared memory to remove this allocation of semaphore.

I tried killing it but of no use as the WP again went into new state. Restart will help. But i need to know the root cause of it

WP 25 had the state "run", has it changed to state "new" after you have tried to kill it? Could you please attached dev_w25 and also dev_w45 (since WP 45 is the longest running WP)? The dpmon output starts with WP 15, in which state are WPs 0 till 14?

Hi karthik,

Follow below thread you find the answer----->

If not find the resolution try this

Restart the Specific instance before you perform----->

clenipc remove <instancenumber> ----with sidadm user

And also check the db related issues.

Regards,

Pravin

Hi Alwina,

The below note helped me. Thanks for your findings.

1795711 - DP: Work processes block on semaphore 42

Hello Karthik,

you are welcome 🙂

Regards,
Alwina

Ask a Question

Top Q&A Solution Author

User

Count