Wednesday, December 14, 2022

EBS -- RDBMS Bugs & Stub Objects remaining even after a full & aggresive online patching cycle that involves full actualization & cleanup

Let's go with the ACS site visits. (still bring us interesting cases) A quick tip from one of the nodes of my professional career. It is almost as if my concentration is in a state of superposition :) Anyways, let's operationalize it!

In  one of my previous blog post, I shared a real life story from my ACS site visits. It was about a full and aggressive online patching cycle that will do well on a loaded obj$. Here is the link for those who want to that -> http://ermanarslan.blogspot.com/2022/11/rdbms-ebs-122-too-many-records-in-obj.html

Now let's add to some extra information ( a crucial tip) to that, in this blog post.

SQL> select sys_context('Userenv','Current_Edition_Name') from dual; 

Although, you may execute a full online patching cycle, that involves full actualization & cleanup;  there may still be stub objects reported..

Example output:

SYS_CONTEXT('USERENV','CURRENT_EDITION_NAME')

V_20221125_1024

SQL> @ADZDSHOWOBJS.sql

Editioned Objects Per Edition

EDITION_NAME       A_VALID  A_INVALID    A_TOTAL    S_VALID  S_INVALID    S_TOTAL      TOTAL

--------------- ---------- ---------- ---------- ---------- ---------- ---------- ----------

V_20210406_1722          0          2          2          6          3          9         11

V_20211022_1201          0          0          0          6          5         11         11

V_20211227_1154          0          0          0          6          5         11         11

V_20220106_1118          0          0          0          6          5         11         11

V_20220112_1427          0          0          0          6          5         11         11

V_20221125_1024     170737        433     171170          5          5         10     171180

Note :

(A_) - Actual Objects 

(S_) - Stub Objects

*Don't misinterpret the situation by the low number of objects here, their reflection on obj$ is huge. In a case where we had 3 million objects in obj$, we implemented that full online patching cycle (actualized all, cleaned up all), and it removed 2 million records in obj$, but! tmore than 500000 records, which should have been also removed, remained.  Well. -> 

*Check the bug database for these/Stubs.. There are one-offs for Linux, but if you are using Solaris or something, you may need to go with PSUs or RUs. (These are evidences of the advantage that we take by being up to date and running Oracle on Linux once again..) Check with ADZDSHOWOBJS.sql and make sure.

*Take action if necessary. A Ref. Bug 32988028 - Stubs Are Not Properly Deleted When an edition is dropped (32988028.8)