Job cancelled after system exception ERROR_MESSAGE in DB13

[Former Member]

Hello All,

When i opened the t-code DB13 i saw that this job "Mark tables requiring statistics update" is cancelled.

JOB LOG:

12.02.2011 22:00:16 Job started

12.02.2011 22:00:16 Step 001 started (program RSDBAJOB, variant &0000000000085, user ID 80000415)

12.02.2011 22:00:18 Job finished

12.02.2011 22:00:18 Job started

12.02.2011 22:00:18 Step 001 started (program RSADAUP2, variant &0000000000081, user ID 80000415)

12.02.2011 22:01:26 Error when performing the action

12.02.2011 22:01:26 Job cancelled after system exception ERROR_MESSAGE

When check for the BGD Job in SM37 for this job i found the same error in the job log with the status cancelled.

Job log overview for job: DBA!PREPUPDSTAT_____@220000/6007 / 22001700

12.02.2011 22:00:18 Job started

12.02.2011 22:00:18 Step 001 started (program RSADAUP2, variant &0000000000081, user ID 80000415)

12.02.2011 22:01:26 Error when performing the action

12.02.2011 22:01:26 Job cancelled after system exception ERROR_MESSAGE

I couldn't find any logs in SM21 of that time also no dumps in ST22.

Possible reason for this error:

I have scheduled the job Check database structure (only tables) at some other time and deleted the earlier job which was scheduled during the bussiness hours which caused performance problem.

So to avoid performance issue i scheduled this job in the mid night by cancelling the old job which was scheduled during the bussiness hours.

And from the next day i could see this error in DB13.

Rest all the backups are running fine but the only job getting cancelled is for "Mark tables requiring statistics update"

Could anyone tell me what should i do to get rid of this error?

Can i schedule this "Mark tables requiring statistics update" again by deleting the old one?

Thanks.

Regards.

Mudassir Imtiaz

[adrian_dorn]

This error happens if Native-SQL calls are not working ==> Have you updated the SAP Kernel software or the DBSL lately? Or perhaps someone changed something in the DB59 integration data ==> Call transaction DB59, perform the connection test and attach the output.

[Former Member]

Hello Adrian,

I tried with Deleting the job and then scheduled it again.

It started working fine.

Could you please tell me, Are these jobs to be scheduled chronologically?

Is there anything like one job has to be followed by another or if we disturb the chronology then such errors are expected.

Thanks for your help buddy.

Regards,

Mudassir.Imtiaz

[adrian_dorn]

>

> Could you please tell me, Are these jobs to be scheduled chronologically?
> Is there anything like one job has to be followed by another or if we disturb the chronology then such errors are expected.

No. You can start the jobs in any order you like. But of course there are some common rules such as e.g. "don't start a backup before a previous backup has finished".

So it is difficult to say why you had this problem. I think that for answering this question it would be necessary that you give us a step-by-step description of what you did so that we can reproduce it and afterwards we should be able to understand what happened.

[Former Member]

Hello Adrian,

Thanks for your response.

Every alternate day we used to have performance issue at 19:00hrs.

Then when i checked what is causing this problem, i discovered that there was a backup "Check Database Structure (tables only)" scheduled at this time and it was mentioned that this backup may cause performance issue.

Then i changed "Check Database Structure (tables only)" the time of this backup to 03:00hrs.

The next day when i checked DB13 i found that one of the backups failed.

i.e. "Mark Tables Requiring Statistics Update"

Then i checked the log which i posted earlier with the error: "Job cancelled after system exception ERROR_MESSAGE"

I posted this error here and then i tried to delete the jobs scheduled i.e. "Mark Tables Requiring Statistics Update" and then re-schedule it at the same time and interval.

And then it started working fine.

So i just got curious to know the cause of the failure of that job.

Thanks.

Regards,

Mudassir.Imtiaz

P.S There is one more thing which i would like to say which is not related to the above issue, and m sorry to discuss this in this thread.

I found a few Bottlenecks in ST04 with Medium and High priority.

Medium: Selects and fetches selectivity 0.53%: 122569 selects and fetches, 413376906 rows read, 2194738 rows qualified.

High: 108771 primary key range accesses, selectivity 0.19%: 402696322 rows read, 763935 rows qualified.

There are a lot these.

I would really appreciate if you tell me what is the cause for these Bottlenecks and how to resolve.

Thanks a lot.

[adrian_dorn]

The Check Database job is well known for causing Performance Problems. Therefore it is recommended to schedule it only at times when few users are in the system, e.g., at night or at the weekend. If the job runs more than 12 hours in your system, I would recommend to schedule it only at weekends.

Another idea: you should check that no other DB13 jobs are started while the Check Database job is still running. Have you checked the start- and end-time of the the Check Database job? Did the Update Statistics perhaps start during this time window?

[Former Member]

Hello Adrian,

I did checked the duration of this job "Check Database" and it lasts not more than 2hrs.

And the job "Marked tables requiring update statistics" was scheduled almost 2hrs after the execution of "Check Database Tables" job.

Could you please help me with the bottlenecks issue in ST04.

Thanks.

Regards,

Mudassir.Imtiaz

[adrian_dorn]

>

> Hello Adrian,
> 
> I did checked the duration of this job "Check Database" and it lasts not more than 2hrs.
> And the job "Marked tables requiring update statistics" was scheduled almost 2hrs after the execution of "Check Database Tables" job

Ok, then I am afraid that I have no further ideas why your UpdateStatistics job did fail. I think we will need to logon directly into your system and debug the UpdateStatistics application to find the reason for the failure. For this it is necessary that your open a ticket in OSS so that we can logon into your system.

>

> Could you please help me with the bottlenecks issue in ST04.

Perhaps you should open a separate topic on this, because probably the guys who are familiar with bottleneck analysis will not look into a topic which has the title "Job cancelled after system exception ERROR_MESSAGE in DB13"

[Former Member]

Hello Adrian,

Sorry for my late response.

Thanks a lot for your support, i will create another thread for the "Bottlenecks Issue".

Regards,

Mudassir.Imtiaz