SeqCALLCENTERCAGLIARICcm has finished, status = 96 (Crashed)

eoyylo · Post by **eoyylo** » Thu Mar 24, 2005 1:40 pm

Hi,
I have a "little" problem with a job sequencer: it runs an another job sequencer and after it abort whith the next error:

LogSeqCALLCENTERCAGLIARICcm..JobControl (DSWaitForJob): Job SeqCALLCENTERCAGLIARICcm has finished, status = 96 (Crashed)

(with a green ballon in the log of director!!!!!!)

The controlled job continue to run and SeqCALLCENTERCAGLIARICcm finish with status: "FINISHED (See Log)" or "FINISHED"

WHY?????????

Can someone tell me why?

Thanks in advance

Regards

Mario

ray.wurlod · Post by **ray.wurlod** » Thu Mar 24, 2005 4:19 pm

Reset the job in Director.

Look for a message in the log called "from previous run...". What does this tell you?

eoyylo · Post by **eoyylo** » Thu Mar 31, 2005 8:57 am

Hi Ray

the message is:

From previous run
DataStage Job 269 Phantom 20751
[20758] Done : DSD.RUN LogH3GITmisACDtoSTGACDUSERSGECcm1. 0/0 DIR_SEMAFORO_GE=/var/opt/ascential/abis/SEMAFORO/CALL_CENTER_GE ACDGEOWNER= ACDGEPSW= ACDGEUSR= ACDGEDB= STGOWNER=ccmetl STGPSW=ccmetl STGUSR=ccmetl STGDB=h3g_prep_ods_rev DSJobController=SeqCALLCENTERGENOVACcm
[20869] Done : DSD.RUN LogH3GITmisACDtoSTGACDAACEVENTGECcm1. 0/0 DIR_SEMAFORO_GE=/var/opt/ascential/abis/SEMAFORO/CALL_CENTER_GE ACDGEOWNER= ACDGEPSW= ACDGEUSR= ACDGEDB= STGOWNER=ccmetl STGPSW=ccmetl STGUSR=ccmetl STGDB=h3g_prep_ods_rev DSJobController=SeqCALLCENTERGENOVACcm
[20835] Done : DSD.RUN LogH3GITmisACDtoSTGACDAACREASONGECcm1. 0/0 DIR_SEMAFORO_GE=/var/opt/ascential/abis/SEMAFORO/CALL_CENTER_GE ACDGEOWNER= ACDGEPSW= ACDGEUSR= ACDGEDB= STGOWNER=ccmetl STGPSW=ccmetl STGUSR=ccmetl STGDB=h3g_prep_ods_rev DSJobController=SeqCALLCENTERGENOVACcm
[20919] Done : DSD.RUN LogH3GITmisACDtoSTGACDCALLDETAILGECcm1. 0/0 DIR_SEMAFORO_GE=/var/opt/ascential/abis/SEMAFORO/CALL_CENTER_GE ACDGEOWNER= ACDGEPSW= ACDGEUSR= ACDGEDB= STGOWNER=ccmetl STGPSW=ccmetl STGUSR=ccmetl STGDB=h3g_prep_ods_rev DSJobController=SeqCALLCENTERGENOVACcm
Program "DSU.ABISMessage": Line 4, Unable to open the operating system file "DSU_BP.O/DsLogWarning".
[ENOENT] No such file or directory
Program "DSU.ABISMessage": Line 4, Unable to load file "DsLogWarning".
Program "DSU.ABISMessage": Line 4, Unable to load subroutine.
[21192] Done : DSD.RUN LogH3GITmisSTGtoDDSTADMAACCREASONGECcm2. 0/0 DIR_SEMAFORO_GE=/var/opt/ascential/abis/SEMAFORO/CALL_CENTER_GE DDSOWNER=ccmbi DDSPSW=ccmbi DDSUSR=ccmbi DDSDB=h3g_prep_ods_rev STGOWNER=ccmetl STGPSW=ccmetl STGUSR=ccmetl STGDB=h3g_prep_ods_rev DSJobController=SeqCALLCENTERGENOVACcm
Attempting to Cleanup after ABORT raised in stage SeqCALLCENTERGENOVACcm..JobControl

DataStage Phantom Aborting with @ABORT.CODE = 3

Sometimes the job sequencer abort with this message of error:

LogH3GITmisDDStoDDSTADMMDETTAGLIOCHIAMATEGECcm2..JobControl (@H3GITmisDDStoDDSTADMMDETTAGLIOCHIAMATEGECcm2): Controller problem: Error calling DSRunJob(H3GITmisDDStoDDSTADMMDETTAGLIOCHIAMATEGECcm2), code=-99

[General repository interface 'other error']

The status of the job was FINISHED (code = 21?, i don't remember) but the sequencer can't "read it".

... I don't understand the mistake

REgards

Mario

mhester · Post by **mhester** » Thu Mar 31, 2005 10:06 am

It would seem to me that there is a user written routine involved here that calls the wrong function. You are calling -

Code: Select all

DsLogWarning

and it should be -

Code: Select all

DSLogWarn

eoyylo · Post by **eoyylo** » Thu Mar 31, 2005 10:40 am

Hi mhester,
the true blocking error are the others:

SeqCALLCENTERCAGLIARICcm has finished, status = 96 (Crashed)

and

LogH3GITmisDDStoDDSTADMMDETTAGLIOCHIAMATEGECcm2..JobControl (@H3GITmisDDStoDDSTADMMDETTAGLIOCHIAMATEGECcm2): Controller problem: Error calling DSRunJob(H3GITmisDDStoDDSTADMMDETTAGLIOCHIAMATEGECcm2), code=-99

[General repository interface 'other error']

The function "DSLogWarning" ( DsLogWarn ) is called after this errors

Regards

Mario

mhester · Post by **mhester** » Thu Mar 31, 2005 11:02 am

Do you set the "Execution Action" to "reset if required, then run" in the job activity stages? I have seen this behavior before when the job is not in a runnable state for whatever reason.

eoyylo · Post by **eoyylo** » Thu Mar 31, 2005 11:11 am

Hi mhester,

the action of all job sequencer is "reset if required, then run".
This error occur whit a new project with all jobs and job sequencer compiled.

I suppose that the problem mybe the configuration of server.

I observed that at same time it is possible that about 90 phantom process and 100 oracle process are running.

Does Datastage have a limit of jobs and job sequencer that could be run in parallel way (on the some project)?

an another information: in the DSHOME/bin director there isn't the dslictool file. Which is its purpose? It is important for this issue?

Regards

mhester · Post by **mhester** » Thu Mar 31, 2005 11:16 am

First, you may be experiencing resource contention so this may/maynot be causing the issue you describe. The -99 or unknown error may indicate just this type of problem. What are the 90 phantom processes?

Second, Ray and Ken may wiegh in with more information, but I believe the dslictool file is not involved with the problem you are experiencing.

Look at post
viewtopic.php?t=91685&highlight=dslictool

Where there is some discussion the dslictool tool.

Regards,

eoyylo · Post by **eoyylo** » Thu Mar 31, 2005 11:23 am

The "phantom process" are the unix process that is runned by Datatage server.
I suppose that
1 phantom unix process = 1 running job

The true problem is that in this Datastage project I have "only" 130 jobs and 150 job sequencer, but in another project I will have about 350 jobs and 400 job sequencer!!!! And all have to run without any problem/ error !!!!

Regards

mhester · Post by **mhester** » Thu Mar 31, 2005 11:34 am

Issues can arise from system resource contention if you are running an extremely large number of processes in parallel. If these are sequential in nature then I would not believe that resource contention would be your problem.

eoyylo · Post by **eoyylo** » Thu Mar 31, 2005 12:18 pm

Can the issue raise because unix server configuration (kernel parameter) or datastage server configuration (uvconfig) aren't correct ?

Which are the parameters of unix kernel that i must verify?

Which are the parameter of uvconfig file that i must verify?

Regards

ray.wurlod · Post by **ray.wurlod** » Thu Mar 31, 2005 4:23 pm

The error is accurately diagnosed by DataStage in the "from previous run" message.

Code: Select all

Unable to load DSLogWarning routine.

It is looking for a routine called DSLogWarning, which exists somewhere in your code.

My guess is that this is in (a Routine called from) the job DSD.RUN LogH3GITmisSTGtoDDSTADMAACCREASONGECcm2, or possibly the job DSD.RUN LogH3GITmisACDtoSTGACDCALLDETAILGECcm1.

The correct DataStage Function for logging warning messages is DSLogWarn, as Michael noted. Find the Routine that invokes DSLogWarning, change that the DSLogWarn, and all should be well.