cancel
Showing results for 
Search instead for 
Did you mean: 

GRC 10.1 SP06 - Role sync fails with dump DBIF_RSQL_INVALID_RSQL

dyaryura
Active Participant
0 Kudos

Hi!

We're experiencing an issue with the repository sync job for one connector. When selecting the full option, the sync for users and profiles report success but for the role sync we get the error "Error in RFC; 'Error in module RSQL of the database interface'" and a dump is generated in the back-end ( DBIF_RSQL_INVALID_RSQL) with the program /GRCPI/CL_GRIA_ROLE===========CP

The problem occurs in the sentence:

>>>>> SELECT agr_name parent_agr FROM agr_define    "#EC CI_SGLSELECT
   84       INTO TABLE lt_derived
   85 *       WHERE agr_name IN it_incl_role
   86       WHERE agr_name IN lt_incl_role
   87          AND ( change_dat GE iv_update_date  OR
   88              ( create_dat GE iv_update_date AND change_dat = '00000000'  ) ).

We've tried working with the parameter 1122 (Batch size for Role sync) switching it to different values like 500, 100, 50, 10 and even 1 but the error still persists.

Synch for another connector with the same GRC System works fine.

any idea?

Thanks,

Diego.

Accepted Solutions (1)

Accepted Solutions (1)

dyaryura
Active Participant
0 Kudos

Hi Guys!

For those who might experience this issue, new note is going to be released soon:

2063384 - Exception CX_SY_OPEN_SQL_DB caught during Role sync or during Repository sync

Cheers,

Diego.

Answers (1)

Answers (1)

Former Member
0 Kudos

Sounds like an old foe from 10.0

http://service.sap.com/sap/support/notes/1632421

You may have to ask SAP to make a 10.1 version of this fix for you to apply to the plug-in system.

There is a specific note for this issue in both 10.0 and 10.1, but it is part of 10.1 SP01 (you are on SP06)

http://service.sap.com/sap/support/notes/1803121

dyaryura
Active Participant
0 Kudos

Hi Harinam!

Yes, I've checked some notes but since we're on the latest SP all of them are already implemented.

We're opening an OSS. I'll update as soon as I have more information.

Thanks!

Diego.

Former Member
0 Kudos

OSS message seems the reasonable (and probably only) way forward.

It would not be the first time some code has been re-altered to cause a bug to re-appear