Job SAP_COLLECTOR_FOR_PERFMONITOR getting delay...

Hi Friends,

In one of my customer system, the job SAP_COLLECTOR_FOR_PERFMONITOR is getting delay on particular time.

Normally every hour it is getting finished with 5 to 40 secs.

But at a particular the job starts at i.e. 02:26am is finishing successfully but taking much time. It's nearly taking 37,000 secs.

Job log:

Job started

Step 001 started (program RSCOLL00, variant , user ID DDIC)

Job finished

My system details are:

Comp version :BBPCRM 4.0

Database system: ORACLE , 10.2.0.4.0.

Kernel release : 640

SAP_BASIS:SAPKB62063

I can submit if any information required. Earlier solution is much aspirated.

Regards

Sudhakar

SAP Managed Tags:
Oracle Database

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)

maybe you have an offline backup running while this job tries to run?

check in SM21 for any activity .

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

Comment

This may be due to all BTC WPs are occupied. You can setup OP modes (RZ03) and change the work process types for day and night. as you can set more BTC for night and more DI for day.

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

Comment

Hi Sunil,

I have checked in the system. BTC is available at acceptable value.

Not much jobs also running at that time.

It seems a bit of clarification is needed here.

What exactly do you mean by delay ?

Two guesses:

1) The job starts later than expected. This is what usually is called delay. And this seems to be how Sunil understood your question.

2) The job starts roughly at the expected time, but the end is delayed. This usually is called long run time. But from your remark 'It's nearly taking 37,000 secs' I gather that this could be what you are talking about.

regards

Hi,

Have you checked st03n to know where is the time spent? or it simply gets hanged doing nothing?

Br,

Javier

Job starts at defined time. its taking much time to get finishes "long run time" & i have checked in sm37 only.

In which tab i can found where it's get delay in st03n.

Job SAP_COLLECTOR_FOR_PERFMONITOR will run every hour, but its task list will vary, that means different reports will be started.

If it has a long run time at a specific hour every day, you should first find out, which of these tasks is taking long.

For a start have a look at:

http://help.sap.com/saphelp_nw70ehp1/helpdata/en/84/9e0e422dfcdc2ce10000000a1550b0/frameset.htm

hope this helps

As per SAP Note 16083 - Standard jobs, reorganization jobs, the SAP_COLLECTOR_FOR_PERFMONITOR job always be scheduled in client 000 with user DDIC or with a user with the same authorization

also go through Note 12103 - Contents of the TCOLL table

Hi,

Have you tried to delete all SAP_COLLECTOR_FOR_PERFMONITOR jobs from the client(s) and re-schedule it again on hourly basis in 000 client ?

Regards,

Bhavik G. Shroff

@ Joe: I have checked the entries of TCOLL in the system all looks same as in DEV & QA environments & the same job is running fine in DEV & QA environments.

@Bhavik : Do you want me to re schedule after cancellation.

& no backup is running at same time.

Hello,

As Francisco Javier stated, in ST03N you can see the programs which runs every execution of SAP_COLLECTOR_FOR_PERFMONITOR. Go to ST03N ( Expert mode) -> Collector and Performance DB ->

Performance monitor collector -> Log.

Here you will see a list of executions like this:

Date Time Status

28.09.2010 18:40:36 No errors

28.09.2010 17:40:35 No errors

28.09.2010 16:40:34 No errors

28.09.2010 15:40:33 No errors

28.09.2010 14:40:32 No errors

28.09.2010 13:40:32 No errors

28.09.2010 12:40:31 No errors

28.09.2010 11:40:30 No errors

28.09.2010 10:40:29 No errors

28.09.2010 09:40:28 No errors

28.09.2010 08:40:27 No errors

28.09.2010 07:40:26 No errors

28.09.2010 06:40:25 No errors

28.09.2010 05:40:25 No errors

28.09.2010 04:40:24 No errors

28.09.2010 03:40:23 No errors

28.09.2010 02:40:22 No errors

28.09.2010 01:40:21 No errors

28.09.2010 00:40:20 No errors

Double click in the corresponding (long) execution and you will see the details of the programs that ran in these Job, with its start time. Then You will have a more clear idea why the runtime was so long.

In table TCOLL you can see the schedule of programs and start times.

Rafa

Hi,

@Bhavik : Do you want me to re schedule after cancellation.

Delete all the instances of same job SAP_COLLECTOR_FOR_PERFMONITOR in the System. then You can reschedule it on hourly basis and monitor its execution.

Also monitor the Background work-processes availability during the time when such delay has seen as in your case. You can set the execution server where Background WPs are highly available.

Regards,

Bhavik G. Shroff

Sorry Guys I was not well to answer your replies.

Thanks for your replies.

Still the issues persists. Here are some more information which would help you to help me.

I have checked BTC are available.

In SM51 one BTC is running for long time for this with a entry of u201CSAPLSTUW 000 DDICu201D

& Daily BTC process is Disconnecting from oracle at 2 Ou2019clk then this job is taking that BTC. Is there any relation between this?

Wed Oct 6 01:50:01 2010

got NLS_LANG='AMERICAN_AMERICA.UTF8' from environment

Client NLS settings:

Logon as OPS$-user to get SAPCRM's password

Connecting as /@SID on connection 1 (nls_hdl 0) ... (dbsl 640 070308)

Nls CharacterSet NationalCharSet C EnvHp ErrHp ErrHpBatch

0 UTF8 1 0x114876050 0x114888db0 0x114888668

Starting user session (con_hdl=1,svchp=0x1160b89c8,srvhp=0x11488bd68,usrhp=0x1160b85b0)

Now '/@SID' is connected: con_hdl=1, nls_hdl=0, session_id=687.

Got SAPCRM's password from OPS$-user

Disconnecting from connection 1 ...

Close user session (con_hdl=1,svchp=0x1160b89c8,usrhp=0x1160b85b0)

Now I'm disconnected from ORACLE

Connecting as SAPCRM/<pwd>@SID on connection 1 (nls_hdl 0) ... (dbsl 640 070308)

Nls CharacterSet NationalCharSet C EnvHp ErrHp ErrHpBatch

0 UTF8 1 0x114876050 0x114888db0 0x114888668

Starting user session (con_hdl=1,svchp=0x1160b89c8,srvhp=0x11488bd68,usrhp=0x1160b85b0)

Now 'SAPCRM/<pwd>@SID' is connected: con_hdl=1, nls_hdl=0, session_id=687.

Connection 1 opened (DBSL handle 1)

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

134 000 R/3 000000000 ACTIVE NO YES NO 000 255 255 20100930 152615 host05

134 001 R/3*WORKFLOWTRAC 134000057 ACTIVE NO NO NO 004 255 255 20101006 015001 host05

ERROR => EmActiveData: Invalid Context Handle -1 [emxx.c 2220]

ERROR => EmActiveData: Invalid Context Handle -1 [emxx.c 2220]

ERROR => EmActiveData: Invalid Context Handle -1 [emxx.c 2220]

Wed Oct 6 02:08:47 2010

Disconnecting from connection 1 ...

Close user session (con_hdl=1,svchp=0x1160b89c8,usrhp=0x1160b85b0)

Now I'm disconnected from ORACLE

Disconnected from connection con_da={R/3*WORKFLOWTRACE,134000057,1}

Job is taking much time to finish at 02:26 with a lock entry in MONI table. In ST03n sometimes 3:26 job is showing error but it is not failing.

Error log:

System name Host name Report Date End time

Error message

Host05_SID_00 Host05 RSCUECRM 06.10.2010 03:26:01

Host05_SID_00 Host05 RSHOSTDB 06.10.2010 03:26:01

Host05_SID_00 Host05 RSORACOL 06.10.2010 03:26:01

Host05_SID_00 Host05 RSORAHCL 06.10.2010 03:26:01

Host05_SID_00 Host05 RSRFCDLT 06.10.2010 03:26:01

Host05_SID_00 Host05 RSRFCDMN 06.10.2010 03:26:01

Host05_SID_00 Host05 RSSTAT60 06.10.2010 03:26:06

Table MONI is locked by user DDIC

Host05_SID_00 Host05 RSSTAT87 06.10.2010 03:26:06

Host05a_SID_01 Host05a RSHOSTDB 06.10.2010 03:26:07

Host05a_SID_01 Host05a RSRFCDMN 06.10.2010 03:26:07

Host05_SID_00 Host05 RSSTAT83 06.10.2010 03:26:11

Begin error recovery:

Host05_SID_00 Host05 RSSTAT60 06.10.2010 03:27:16

Table MONI is locked by user DDIC

Contents of TCOLL

Report Days of Week Hours of Day Repe- System

titions

Mo We Fr Su 0 2 4 6 8 10 12 14 16 18 20 22

RSCUECRM X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X 1 C

RSDBPREV X X X X X X X X X X X X X X X X X X X X 1 C

RSDB_PAR X X X X X X X X X X 1 C

RSDB_TDB X X X X X X X X X 1 C

RSDB_WDB X X 1 C

RSHOSTDB X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X 1 *

RSHOSTDC X X X X X X X X X X X X X X X X X X X 1 C

RSHOSTPH X X X X X X X X X X 1 *

RSORA811 X X X X X X X X X 1 C

RSORACOL X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X 1 C

RSORAHCL X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X 1 C

RSRFCDLT X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X 1 C

RSRFCDMN X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X 1 *

RSSDBDLY X X X X X X X X 1 C

RSSTAT60 X X X X X X X X X X X X X X X 1 C

RSSTAT83 X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X 1 C

RSSTAT87 X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X 1 C

RSSTAT90 X X X X X X X X X 1 *

RSSTAT98 X X X X X X X X X X X X X 1 *

RSSTATPH X X X X X X X X X X 1 *

RSTUNE80 X X X X X X X X X X 1 *

RSXRPM X X 1 C

You showed the ST03N entries for the 3:26 job.

But I think the entries for the 2:26 job would be more helpful.

We want to find out why the 2:26 job is slow.

We don't want to find out why the 3:26 job is fast.

regards

Here are yesterdays 02:26 job log. Todays job is still running.

System name Host name Report Date End time

Error message

Host05_SID_00 Host05 RSCUECRM 05.10.2010 02:26:01

Host05_SID_00 Host05 RSDBPREV 05.10.2010 02:26:01

Host05_SID_00 Host05 RSHOSTDB 05.10.2010 02:26:01

Host05_SID_00 Host05 RSHOSTDC 05.10.2010 02:26:01

Host05_SID_00 Host05 RSORACOL 05.10.2010 02:26:01

Host05_SID_00 Host05 RSORAHCL 05.10.2010 02:26:01

Host05_SID_00 Host05 RSRFCDLT 05.10.2010 02:26:01

Host05_SID_00 Host05 RSRFCDMN 05.10.2010 02:26:01

Host05_SID_00 Host05 RSSTAT87 05.10.2010 02:26:01

Host05a_SID_01 Host05a RSHOSTDB 05.10.2010 02:26:01

Host05a_SID_01 Host05a RSRFCDMN 05.10.2010 02:26:01

Host05_SID_00 Host05 RSSTAT83 05.10.2010 04:28:37

> Hi, > >

@Bhavik : Do you want me to re schedule after cancellation.

> Delete all the instances of same job SAP_COLLECTOR_FOR_PERFMONITOR in the System. then You can reschedule it on hourly basis and monitor its execution.

> Also monitor the Background work-processes availability during the time when such delay has seen as in your case. You can set the execution server where Background WPs are highly available.

> Regards,

> Bhavik G. Shroff

Hi,

The job SAP_COLLECTOR_FOR_PERFMONITOR is already scheduled on hourly basis, how would rescheduling it again on hourly basis improve the job runtime?

Thank you.

Regards,

SAP BASIS Trainee

Ask a Question

Top Q&A Solution Author

User

Count