on 08-07-2010 2:18 AM
Hi Experts,
Do you have any idea why we get this error in MDS log file? I am sure you must have come across it, what could be a fix for it?
We have observed Log info "Sort Index Size failed in file" logged into MDS Log file while doing following:
It happens during following processes:
a)_Master Slave Sync between Master and Slave Repo->> MDSAdminServer@GetAndApplyMasterChangeRep_
2010-08-05T12:00:01.012 14 23 Condition: pos < sortIndexSize failed in file /export/home/builder/mdm.build/7.1.03.xx/Cpp/MDM/Servers/MDS /DataMgmt/Update.cpp , line 1097
MDSAdminServer@GetAndApplyMasterChangeRep Dbg.cpp 65
Admin SRM_MDM_CATALOG_CHILDPROD3-DB- HA:1527/PMDORCL_21_20_4_3
b)_Remove Records from Data Manager->> CatMgr@RemoveRecords336_
2010-07-30T21:00:02.378 42 23 Condition: pos < sortIndexSize failed in file /export/home/builder/mdm.build/7.1.03.xx/Cpp/MDM/Servers/MDS/DataMgmt/Update.cpp , line 1097
MDSAdminServer@GetAndApplyMasterChangeRep Dbg.cpp 65
Admin SRM_MDM_CATALOG_CHILDPROD3-DB- HA:1527/PMDORCL_21_20_4_3
c)_Update Records from Data Manager->> CatMgr@ModifyRecords326_
2010-08-04T22:24:11.602 43 23 Condition: pos < sortIndexSize failed in file /export/home/builder/mdm.build/7.1.03.xx/Cpp/MDM/Servers/MDS/DataMgmt/Update.cpp , line 1097 CatMgr@RemoveRecords336 Dbg.cpp 65 49 harrik3 SRM_MDM_CATALOG <PROD3-DB-HA:1527/PMD[Oracle]>
As the Log Info name indicates "Sort Index Size failed in file", we are wondering is there anything to do with the -Size of Index File. Is there any parameter in DB side or in MDS which controls logging file size. May be that file size is not allowing Index sorting to be written into that file and file gets overflown due to less file size and MDS Logs this "Log Info". And since Indexing is not completed and not written into Index file, eventually the search made thru Advance Search gets weird and start throwing null pinter exception in browser because Index is not present for some searches.
This error seems like some unhandled exception code in program- Update.cpp at line 1097.
I would appreciate if you can throw some light on this and how to avoid this?
Regards
Pravin
Hi Pravin,
Just check for below SAP Note's:
SAP Note 1342611: Fixed: Sort index changes were not reflected at the Search pane for a lookup field in a qualified lookup table
SAP Note 1385073: Fixed: Loading specific repository with undefined sort index caused server crash on UNIX
Regards,
Mandeep Saini
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Hi Pravin,
Please check installation guide on page 44,47-51.
Hope it helps you.
Thanks,
Sudhanshu
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Hi Pravin
Not sure on this. can you please check what is the number of partitions selected when you create a new slave. Copying from the ref guide-
A repository partition is simply the underlying physical database in which the repository is stored. By default, MDM gives a new MDM repository the number of partitions currently set for the Default Partitions DBMS setting (one, two, or four).
You can change the number of partitions by selecting the applicable radio button in the Repository Options dialog.
hope this helps-Ravi
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Hi Pravin,
After reviewing MDS log details i think you try to sync master and slave repository try out following option
1. have you did MDM version upgradation if yes then need to create new slave repository
2. check the icon of your slave repository if icon is broken then repository corrupted need to rebuild slave repository
3. if you have done any structural modification to repository then load repository with update indices then sync slave repository
4. verify your repository if it has error then repair/load master repository and sync slave repository
Thanks,
Jignesh Patel
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Hi Jignesh and Ravi,
There is no Repo schema change and we made sure that we create new slave after any Repo schema change. I dont think there is an issue with master-slave, but issue probably lies some where else.
In fact my original question is scrambled. Let me mention it again clearly. Its not alone master-slave sync we get this log info-" Sort Index Size failed in file". Actually we get it during following 3 operations:
a) On Master Slave Sync between Master and Slave Repo->> MDSAdminServer@GetAndApplyMasterChangeRep
b) OnRemove Records from Data Manager in master->> CatMgr@RemoveRecords336
c) On Updating Records in Data Manager->> CatMgr@ModifyRecords326
Is there anything to do with the -Size of Index File. Is there any parameter in DB side or in MDS which controls logging file size. This error seems like some unhandled exception code in program- Update.cpp at line 1097.
_Pravin
User | Count |
---|---|
87 | |
10 | |
10 | |
10 | |
7 | |
6 | |
6 | |
5 | |
5 | |
4 |
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.