Job status...return value...stange problem...

[kaps]

I have a strange problem. We run the Datastage jobs from a scheduling tool using unix script. Few of my jobs failed due to No space left on device problem but it looks like status sent to scheduling tool is (0) success. The actual status is failed.

Does anyone faced such problem ? Does it return any other code when the job fails for space reason ?

we check the status of the job using the following command

dsjob -run -jobstatus $prmStr $prm_project $prm_job

Following steps are done afterwords...
Getting end event ID
Sending the log to a file
Check the status and exit
Is there any other way I need to check for jib status ?

Please let me know...
[/quote]

[ray.wurlod]

dsjob -jobinfo $prm_project $prm_job

[kaps]

Ray

First job did not run but the scheduler moved on to next job which failed because the input file was not there. First job is the one which failed because of the space issue. But later we reran the jobs to completion.
So now If I check the status it will give me the return status of success.

I am trying to figure out why it was sending success last night even though it was in the failed status. is there a way we can check the job status of previous run ?

Thanks

[chulett]

With the -jobstatus option, a return value of 1 is 'success', not zero. An abort would have returned a 3.

[kaps]

Craig

It's just a mistake. we are actually looking for 1.
Following is the code we are using...

Code: Select all

Run_Stat=`dsjob -run -jobstatus $prmStr $prm_project $prm_job 2> /dev/null`
Run_Stat_Val=`echo $Run_Stat | head -1 | cut -d"(" -f2 | cut -d")" -f1`

if test $Run_Stat_Val -eq 1
then
Ret_Val=0
 echo "DataStage Job $prm_job ran successfully" >> $Script_LogDir/$Script_LogFileName
else
Ret_Val=$Run_Stat_Val
 echo "DataStage Job $prm_job failed...Aborting the script" >> $Script_LogDir/$Script_LogFileName
fi

[rleishman]

Do you need a -wait to wait for a result?

[chulett]

Nope, the -jobstatus option includes the -wait, no extra charge.