10-07-2008 10:14 AM
Since our upgrade that is ECC 6.0/unix we have some issues like some of the users profile reports are producing strange results, for example in suim if I see who has access to MM02 I get a list of 107 people however if I logon as user Test1 and try mm01 the system errors with no authorization.( i have checked the validity period) If we then save the user and rerun suim transaction the user is removed from the list.
Is there away of savings all user profiles .
Can anyone help me please.
thanks
anur
10-07-2008 3:07 PM
This could also be caused by the UST* tables (used mostly by SUIM) becoming out-of-sync with the USR* tables (used formerly but now less by the authority-check itself), however this typically provides more access than that which SUIM reports.
If you can make this same observation in your QA system, then execute transaction SUIM_OLD to see whether problem goes away one-shot as test first.
What SUIM_OLD does is:
IF NOT sync_u IS INITIAL.
CALL FUNCTION 'SUSR_SYNC_USER_TABLES' IN BACKGROUND TASK
EXPORTING
tabletype = 'X'.
* CLIENT =
* exceptions
* bad_client = 1
* others = 2.
IF sy-subrc <> 0.
* MESSAGE ID SY-MSGID TYPE SY-MSGTY NUMBER SY-MSGNO
* WITH SY-MSGV1 SY-MSGV2 SY-MSGV3 SY-MSGV4.
ENDIF.
ENDIF.
.... which you can also do manually from tcode SE37 etc.
Cheers,
Julius
10-07-2008 10:20 AM
Hi Anuroop,
Could you please check if the profiles have been generated properly in the system or not?If not, then you will need to generate the profiles, save it to a transport and then move it across your landscape.
Thanks,
Saby..
10-07-2008 10:46 AM
Hi,
How should i check whether the profiles are generated properly or not.
thanks
10-07-2008 10:47 AM
> How should i check whether the profiles are generated properly or not.
Transaction SUPC will help you there.
10-07-2008 10:47 AM
I think the user comparison may be your issue. (since saving a user solved the problem for that userid)
Run transaction PFUD and reschedule PFCG_TIME_DEPENDENCY.
10-07-2008 3:07 PM
This could also be caused by the UST* tables (used mostly by SUIM) becoming out-of-sync with the USR* tables (used formerly but now less by the authority-check itself), however this typically provides more access than that which SUIM reports.
If you can make this same observation in your QA system, then execute transaction SUIM_OLD to see whether problem goes away one-shot as test first.
What SUIM_OLD does is:
IF NOT sync_u IS INITIAL.
CALL FUNCTION 'SUSR_SYNC_USER_TABLES' IN BACKGROUND TASK
EXPORTING
tabletype = 'X'.
* CLIENT =
* exceptions
* bad_client = 1
* others = 2.
IF sy-subrc <> 0.
* MESSAGE ID SY-MSGID TYPE SY-MSGTY NUMBER SY-MSGNO
* WITH SY-MSGV1 SY-MSGV2 SY-MSGV3 SY-MSGV4.
ENDIF.
ENDIF.
.... which you can also do manually from tcode SE37 etc.
Cheers,
Julius
10-07-2008 3:48 PM
Hi,
Thanks for the reply. I understood but i really i could n't get you.
Thanks
10-07-2008 4:46 PM
What I am saying is that there might be some data in the UST* tables which is "stranded" from the profile data used by the user management system.
When saving in SU01 it appears to be removing these "stranged" or "inconsistent" entries.
If you run SUSR_SYNC_USER_TABLES for table type 'X' it should remove these.
Cheers,
Julius
10-08-2008 1:58 PM
Hi
I have tried as you said but still the problm is not resolved . i even tried PFUD but no success.
thanks
10-08-2008 2:17 PM
Other approaches are:
- Immediately after the failed start of MM02, enter /nSU53 in the ok-code to see whether it really is the S_TCODE check which is failing. See SE93 to the additional checks at t-code start...
- Debug the SUIM report and place a watch-point on S_TCODE for when it's value is MM02 or the 'BIS' field has any value, as MM02 might be included in a range.
... but still, the answer must have something to do with saving once in SU01.
Cheers,
Julius
Edited by: Julius Bussche on Oct 8, 2008 3:18 PM
10-10-2008 11:10 AM
It is weird, i have assigned same roles and the same profiles to the users but only one user is display when i try to check in SUIM.
I have checked profile, validity date etc.
Please help
10-10-2008 4:36 PM
10-15-2008 5:21 PM