UJBR Backup failure_STRING_SIZE_TOO_LARGE

[Former Member]

Hi,

We are on SAPKW73010,SAPK-80010INCPMBPC.

We take UJBR back up every week.

But this week it is failed with a short dump, run time error being "STRING_SIZE_TOO_LARGE".

Surprisingly there is no growth in the data for last 2-3 weeks.Last week we haven't faced this issue.

1805490 - Fail to backup environment with huge amount transaction data

May not be applicable,as we are already on SP10.

I have checked manual activity in this note and those parameters already configured in the system.

Any more suggestions,please?

Thanks,

Manohar

[former_member186338 · Accepted solution]

Dump screen?

Vadim

[Former Member · Accepted solution]

Hi Vadim,

This is the screenshot.

1185658455 is the length of the file(For masterdata,meta data last week.)

Don't know how the size is doubled this time.

Thanks,

Manohar

[former_member186338 · Accepted solution]

1. Try light optimization before backup

2. Check that manual steps for the note are completed.

3. Upgrade to recent SP, SP10 is Old!

Vadim

[Former Member · Accepted solution]

Hi Vadim,

LO was done before Back up job.

Those manual steps are completed.As we are are on support pack,it was done by default I think.

Till last week it was fine.

And there is nu much data added to the system during this week.

Thanks,

Manohar

[former_member186338 · Accepted solution]

"I think." Always check!

But you are talking about significant size increase... May be some incorrect script was run...

Try to understand what was the reason for size increase!

[Former Member · Accepted solution]

Hi Vadim,

I have checked those parameters and they are correct.

I think I got the answer.Will update you tomorrow.

The person executed 1 back up with UJBR T code and other back up with a process chain.

This may be the reason.Today night again will execute either with process chain or with UJBR and let you know the result.

Thanks,

Manohar

[Former Member · Accepted solution]

Hi,

Executed UJBR back up with 2 changes.

1.Packaged by million records changed to 20 from 25.(Few times we ran with 100 also successfully.)

2.Un checked DM data files.

It went well.

I suspect DM data files was culprit.Will isolate the issue.

Thanks,

Manohar