Oracle patching issues

 

Before starting Patching 
-- Take Backup 
-- Run pre check 

-- run analyze -- [root@node1 ~]$ opatchauto apply /u01/patches/30501910 -analyze
--  run lsinventory first  if new opatch 

-- run conflict check first 

-- check invalid  objects 

After Patching 

-- run post checks to  check inventory details and invalid objects 

Frequent solutions  used 

1) Change permissions to 660  

[root@rac01]# find /u01/app/19.8.0/grid_1 -user root -exec chown grid {} \;

2) root@racdbnode02~ # opatchauto resume 

3) opatch -rollback -id 

Ways to improve patching speed 

Purge recycle bin  in database .

Use Out-of-Place Patching: Install the new base release and apply patches to a brand-new Oracle Home first, then switch the active home (-switchGridHome) to minimize active downtime. [1, 2]

Leverage Oracle AutoUpgrade Distributed PDB Patching: For multi-tenant container databases in a RAC environment, enable distributed patching via the parameter tune_setting with distributed upgrade = true to spread PDB upgrades across multiple cluster nodes in parallel. [1]

Update OPatch First: Always download and use the latest version of the OPatch utility in your new home to prevent slow inventory checks from dragging out the pre-patch phase. [1, 2, 3]

Perform Rolling Upgrades Efficiently: Patch cluster nodes one at a time using rolling mode, but tune your service drain timeouts appropriately (avoiding values that are too short to cause login storms or too long to leave the cluster at half capacity)

 

ORA-20001  and ORA-20008

Queryable Patch Inventory - Issues/Solutions for ORA-20001: Latest xml inventory is not loaded into table (Doc ID 1602089.1)
verify_queryable_inventory returned ORA-20001: Latest xml inventory is not loaded into table

select dbms_sqlpatch.verify_queryable_inventory from dual;
VERIFY_QUERYABLE_INVENTORY
--------------------------------------------------------------------------------
ORA-20001: Latest xml inventory is not loaded into table

alter system set "_bug27355984_xt_preproc_timeout"=1000 scope=spfile ;
alter system set "_enable_ptime_update_for_sys"=TRUE   scope=spfile ;

After patching :
alter system reset  "_bug27355984_xt_preproc_timeout"  scope=spfile ;
alter system reset  "_enable_ptime_update_for_sys"   scope=spfile ;

or 

Database administrators can successfully resolve the Oracle 19c datapatch ORA-20001 error, which signals an incomplete patch application, by modifying the qopiprep.bat script’s shell interpreter and then re-running the datapatch utility.

• The appearance of ORA-20001: Latest XML inventory is not loaded into the table indicates that a recent patch was likely applied to the binaries, but the required SQL changes via datapatch either failed or were skipped during execution.
• The first step in fixing the issue requires backing up the relevant script by copying it $ORACLE_HOME/QOpatch/qopiprep.bat to qopiprep.bat_bkp.
• To resolve the script failure, modify the qopiprep.bat file by changing the shell interpreter line from #!/bin/sh to #!/usr/bin/ksh.
• After modifying the script, execute the ./datapatch -verbose command from $ORACLE_HOME/OPatch, and confirm success using the dbms_sqlpatch.verify_queryable_inventory query.
• 
  

Similarly if ORA-20008:  is reported we need use below trace 
>> Before patching 
SQL: alter session set events '18219841 trace name context forever';

>> After patching 
alter session set events '18219841 trace name context off';

1. OPatch Version Mismatch

Error

OPatch version is too old for this patch

or

OPatch version check failed

Cause

Your OPatch version does not meet the minimum requirement of the patch.

Fix

Check current OPatch version:

$ $ORACLE_HOME/OPatch/opatch version

Download latest OPatch from Oracle Support → Patch 6880880.

Replace OPatch directory:

$ mv $ORACLE_HOME/OPatch $ORACLE_HOME/OPatch_old

$ unzip p6880880*.zip -d $ORACLE_HOME

Recheck:

$ opatch version

2. OPatch Cannot Find OUI Inventory

Error

OUI-67124: Inventory is corrupted

or

lsinventory cannot open Oracle inventory

Cause

The Oracle inventory is misplaced, corrupted, or incorrectly referenced.

Fix

Check inventory location:

$ cat /etc/oraInst.loc

Update correct location:

inventory_loc=/u01/app/oraInventory

inst_group=oinstall

Re-run:

$ opatch lsinventory

If corrupted:

$ opatch lsinventory -detail -invPtrLoc /etc/oraInst.loc

3. Missing Required Privileges

Error

Prerequisite check failed

User does not have required permissions

Cause

OPatch must be run by the correct OS user.

Fix

For database home → run as oracle

For GI home → run as grid

Check permissions:

$ ls -ld $ORACLE_HOME

Correct them if needed:

$ chown -R grid:oinstall $GRID_HOME

4. OPatch Cannot Access Files / Permission Denied

Error

Permission denied accessing file ...

Cause

Incorrect ownership or permissions inside Oracle home.

Fix

$ chown -R oracle:oinstall $ORACLE_HOME

$ chmod -R 775 $ORACLE_HOME

For GI:

$ chown -R grid:oinstall $GRID_HOME

5. Patch Conflict Detected

Error

Patch conflict detected with patch <patch_number>

Cause

Another patch is already applied that conflicts with the new one.

Fix

Check applied patches:

$ opatch lsinventory

If rollback is allowed:

$ opatch rollback -id <conflicting_patch>

Reapply desired patch:

$ opatch apply

If conflict is mandatory → download merged patch from Oracle Support.

6. OUI-67074: Component Dependencies Not Satisfied

Cause

You are trying to apply patch on incomplete or partially installed home.

Fix

Validate home:

$ opatch lsinventory -detail

Relink Oracle binaries:

$ $ORACLE_HOME/bin/relink all

Retry patch.

7. OPatch Failed with Java Errors

Error

Java could not be launched

Cause

Missing Java or corrupted ORACLE_HOME Java directory.

Fix

Ensure Java exists:

$ ls $ORACLE_HOME/jdk/bin/java

If missing → copy from same-version home or reinstall OPatch.

8. GI Patching – CRS-1705 / Resources Must Be Stopped

Error

CRS-1705: Resource is not offline

Cause

Required cluster resources are still online.

Fix

Put cluster in correct state:

$ grid_home/bin/crsctl stop crs

or

$ grid_home/bin/crsctl stop cluster -all

Use:

$ opatchauto apply <patch>

9. OPatch Fails with “Out of Space”

Error

No space left on device

Fix

Check filesystem:

$ df -h

Clear temp:

$ rm -rf /tmp/*

$ rm -rf /u01/app/oracle/product/patch_backups/*