MEMORY ALLOCATION FAILURE when running the R30835

[tlwalker3]

I have Oracle Support churning on this one, but decided to post the issue here to see if anyone else is having this issue. We are trying to update the standards on some 4000+ items. We were trying to update about 1000 items (using a cat code value on F4102 for data selection). The report runs for many hours and ends in ERROR. UBE logs are shown below. We submitted it again with a smaller data selection and monitored the batch server's memory. The memory consumption for the job is constantly growing with no commit or memory flush. We suspect this is the reason for the failure. Anyone else experience this issue, and what was your solution?

MEMORY ALLOCATION FAILURE
File: ..\..\Common\jdbdrv\DBSQL\DBSQL_X.C Line:26 iParam: 0000008412

MEMORY ALLOCATION FAILURE
File: ..\..\Common\jdbdrv\DBSQL\DBSQL_X.C Line:26 iParam: 0000008412

MEMORY ALLOCATION FAILURE
File: ..\..\Common\jdekrnl\JDB\JDB_UTL1.C Line:1288 iParam: 0000020000

MEMORY ALLOCATION FAILURE
File: ..\..\Common\jdekrnl\JDB\JDB_UTL1.C Line:1288 iParam: 0000020000

MEMORY ALLOCATION FAILURE
File: ..\..\Common\jdekrnl\jdecache\jdecm_rb.c Line:218 iParam: 0000002116

Configuration:
E1 9.0 Tools 8.98.4.2
E1 Server: Win 2008R2 with 24 Gig Memory
DB Server: SQL2008 with 128 Gig Memory

Thanks
Tim

 

[Larry_Jones]

Hey Tim,

Long, long ago ... on a release far far away ...

We had similar issues with the program under 8.9.
Below is our change log history for how we dealt with the issue. Our speculation was that the specs were corrupt for the object but ...

04/25/2005 Had to recreate call to R30834 in End Section of section S30835E to stop program ending with memory violation
05/16/2005 Deleted bad commented-out call to R30834 in End Section of section S30835E to stop program ending with memory violation

 

[tlwalker3]

Larry,
Thank you for the info. I will get with our development group and see if we can test this. This really looks like some poorly written code, not that I'm saying Oracle would ever write bad code. :0)