Oracle 19c datapatch problem

From: Mladen Gogala <gogala.mladen_at_gmail.com>
Date: Thu, 16 Jul 2020 19:11:32 -0400
Message-ID: <420ceec8-1b6f-14fa-9f64-5af47f3939d2_at_gmail.com>



I've installed July 2020 RU to one of my databases and have had a heck of a time to figure it out. I was installing the RU on top of Oracle 19.7 EE. The "opatch apply" went without a hitch but datapatch -verbose was failing with the following message:

catconInit: unexpected error in get_instance_status_and_name

The log files were as useless as useless can be, without the least indication as to what may be the problem. I sometimes think that Oracle developers really hate database administrators and want to exact revenge on them for the sins committed by our forefathers, somebody like the last common ancestor of Tom Kyte and me. I am sure that such a creature will one day be found somewhere in the Great Rift Valley. However, Oracle Support saved the day:

Datapatch fails with catconInit: unexpected error in get_instance_status_and_name (Doc ID 2409301.1)

It turns out that modified glogin.sql can break the datapatch utility. People will soon start installing this around and glogin.sql is a very frequently modified file. So, if you don't want to hit the same problem, here are the cause and the solution:

    Cause

The issue was caused by customization to the <ORACLE_HOME>/sqlplus/admin/glogin.sql

The datapatch script connects to the database to apply the patch and it expects a certain response to verify that it has connected, but since the glogin was customized datapatch does not understand the response and so it fails.

    Solution

  1. Temporarily rename <ORACLE_HOME>/sqlplus/admin/glogin.sql
  2. Retry datapatch
  3. Rename glogin.sql back to the original name
-- 
Mladen Gogala
Database Consultant
Tel: (347) 321-1217


--
http://www.freelists.org/webpage/oracle-l
Received on Fri Jul 17 2020 - 01:11:32 CEST

Original text of this message