Critical Patch 19c April 2026

   

Oracle Critical Patch Update on 19c - April 2026

GI RU Combo:

Patch 39062956 - Combo of OJVM Component Release Update 19.31.0.0.260421 + Grid Infrastructure Apr 2026 Release Update 19.31.0.0.260421 (REL-APR2026)

PURPOSE: All documents are provided on this Blog just for educational purposes only.  Please make sure that you run it in your test environment before to move on to production environment. 

Execute the following commands:

STEP 1

GI RELEASE UPDATE 19.31.0.0.0 (REL-APR2026)

1)

NOTE: IMPORTANT

GI Home:

You need to update first opatch to the higher or latest version.

You can download latest version from the following link.

For April CPU is required OPatch latest version. It can be downloaded by the following link

Download lest OPatch from "https://updates.oracle.com/download/6880880.html"

Latest OPatch copy files.

p6880880_190000_Linux-x86-64.zip

Grid_home

Als root user:

cd /oracle19/gridbase/grid

unalias mv

cd /oracle19/gridbase/grid

mv OPatch OPatch_OLD

cd $DOWNLOAD_HOME/OPatch_12.2.0.1.51

cp -r  OPatch /oracle19/gridbase/grid

cd /oracle19/gridbase/grid

chown -R oracle:dba OPatch/

su - oracle

. set_home_env +ASM

$ORACLE_HOME/OPatch/opatch version

You Need to do the same on ORACLE_HOME

Oracle ORACLE_HOME

Latest OPatch copy files

Oracle_home:

Als root user:

cd /oracle19/support/download/OPatch_12.2.0.1.51

cd /oracle19/dbbase/db

unalias mv

mv OPatch OPatch_OLD

cd /oracle19/support/download/OPatch_12.2.0.1.51

cp -r  OPatch /oracle19/dbbase/db

cd /oracle19/dbbase/db

chown -R oracle:dba OPatch

su - oracle

. set_home_env ${ORACLE_SID}

$ORACLE_HOME/OPatch/opatch version

Unzip April 2026 combo patch file downloaded.

p39062956_190000_Linux-x86-64.zip

Check the conflict:

From Grid_Home

$DOWNLOAD_PATCH is a directory where the patch is located.

$ORACLE_HOME/OPatch/opatch prereq CheckConflictAgainstOHWithDetail -phBaseDir  $DOWNLOAD_PATCH/39062956/39036936/39034528

$ORACLE_HOME/OPatch/opatch prereq CheckConflictAgainstOHWithDetail -phBaseDir $DOWNLOAD_PATCH/39062956/39036936/39039430

$ORACLE_HOME/OPatch/opatch prereq CheckConflictAgainstOHWithDetail -phBaseDir $DOWNLOAD_PATCH/39062956/39036936/39055473

$ORACLE_HOME/OPatch/opatch prereq CheckConflictAgainstOHWithDetail -phBaseDir $DOWNLOAD_PATCH/39062956/39036936/39107855

$ORACLE_HOME/OPatch/opatch prereq CheckConflictAgainstOHWithDetail -

phBaseDir $DOWNLOAD_PATCH/39062956/39036936/39107825

For Database home, as home user

$ORACLE_HOME/OPatch/opatch prereq CheckConflictAgainstOHWithDetail -phBaseDir $DOWNLOAD_PATCH/39062956/39036936/39034528

$ORACLE_HOME/OPatch/opatch prereq CheckConflictAgainstOHWithDetail -phBaseDir $DOWNLOAD_PATCH/39062956/39036936/39039430

2)

Shut down all instances and listeners associated with the Oracle home that you are updating.

NOTE:

If you have standby DB. Stop them first and keep it for fallback:

On Standby:

select name, open_mode, database_role from v$database;

select process, status, sequence# from v$managed_standby;

alter database recover managed standby database cancel;

On Primary:

show parameter log_archive_dest_state_2;

alter system set log_archive_dest_state_2=DEFER SCOPE=BOTH;

On Standby:

 shutdown immediate;

 srvctl stop asm -f

On Primary:

STOP DB and PDB from Oracle home:

alter pluggable database KARDBPDB close immediate;

srvctl stop database -d KARDBTST

srvctl status database -d KARDBTST

Database is not running.

Stop Listener from GI Home

srvctl stop  listener -l LISTENER

srvctl status listener -l LISTENER

Listener LISTENER is enabled

Listener LISTENER is not running

Stop ASM

srvctl stop asm -f

srvctl status asm

ASM is not running.

3)

As root user execute the following command.

You have following options

 i) To patch only the GI Home

 <GI_HOME>/OPatch/opatchauto apply <UNZIPPED_PATCH_LOCATION>/patchid -  oh <GI_HOME>

ii) To patch oracle_home

  opatchauto apply <UNZIPPED_PATCH_LOCATION>/patchid -oh <oracle_home1_path>

 iii) Patching Oracle Home and Grid Home togather:

 As root user, execute the following command on each node of the cluster:

 # <GI_HOME>/OPatch/opatchauto apply <UNZIPPED_PATCH_LOCATION>/patchid

I am going to apply the path for both grid home and oracle home together.

Oracle Document: 

https://updates.oracle.com/Orion/Services/download?type=readme&aru=28686071

NOTE: make sure that the Clusterware has is running:

crsctl status has

If it is not running, you need to start it as follow.

crsctl start has

As root user:

cd /users/oracle

. set_home_env +ASM

export PATH=$PATH:$GRID_HOME/OPatch
$GRID_HOME/OPatch/opatchauto apply $DOWNLOAD_PATCH/39062956/39036936

As Grid_Home:

crsctl start res -all

or

srvctl start asm

srvctl start database -db ${ORACLE_SID} 

Post-Installation Instructions

Single/Multitenant (CDB/PDB) DB

sqlplus / as sysdba

startup

alter pluggable database all open;

quit

cd $ORACLE_HOME/OPatch

./datapatch -verbose

Any databases that have invalid objects need to execute utlrp.sql run

For example:

cd $ORACLE_HOME/rdbms/admin

sqlplus /nolog

SQL> CONNECT / AS SYSDBA

SQL> @$ORACLE_HOME/rdbms/admin/utlrp.sql

-- You can check valid or invalid objects with the following command.

set pagesize 250;

col STATUS format a20;

col COMP_NAME format a35;

select COMP_NAME, version, status from dba_registry;

STEP 2

OJVM RELEASE UPDATE 19.31.0.0.0

For a Non Oracle RAC environment, shut down all databases and listeners associated with the Oracle home that you are updating.

From Oracle_home.

Stop DB - no need to stop ASM

alter pluggable database KARDBPDB close immediate;

srvctl stop database -d KARDBTST

srvctl stop  listener -l LISTENER

export PATH=$PATH:/usr/ccs/bin

#$ORACLE_HOME/OPatch/opatch apply $DOWNLOAD_PATCH/39062956/38906621

Post installation Steps.

startup upgrade

alter pluggable database all open upgrade;

exit

cd $ORACLE_HOME/OPatch

./datapatch -verbose

sqlplus / as sysdba

shutdown

startup

alter pluggable database all open;

SQL> @$ORACLE_HOME/rdbms/admin/utlrp.sql

NOTE: If need Roolback the patch:

 opatch rollback -id 38906621

startup

cd $ORACLE_HOME/OPatch

 ./datapatch -verbose

## Roolback End --

==== Patch Apply DONE On Primary DB====

Verify the patch Version and Status of the patch:

col ID format a5

col COMMENTS format a20

col VERSION format a12

col BUNDLE format a5

col ACTION_TIME format a30

col ACTION format a10

set linesize 150

select substr(action_time,1,30) action_time, substr(id,1,8) id, substr(action,1,10) action, substr(version,1,8) version, substr(comments,1,20) comments from sys.registry$history ;

set pagesize 20;

set linesize 200;

col ACTION_TIME format a30;

col DESCRIPTION format a50;

select PATCH_ID,ACTION,ACTION_TIME,DESCRIPTION,STATUS from registry$sqlpatch order by ACTION_TIME ;

select PATCH_ID,DESCRIPTION,STATUS from registry$sqlpatch order by ACTION_TIME ;

OR

GRID_HOEM:

$ORACLE_HOME/OPatch/opatch lspatches -oh ${ORACLE_HOME}

ORACLE_HOME

$ORACLE_HOME/OPatch/opatch lspatches -oh ${ORACLE_HOME}

Patch Apply on Standby DB - Just in case if you have one:

Aapply a patch you might need to execute ONE or TWO steps

== > Patch Oracle binaries < == On BOTH databases : Primary AND Standby

== > Execute post patching step(s) < == On the Primary ONLY

== > A Physical Standby can NOT be modified DIRECTLY == > Only by applying changes received from the Primary database via redo log files

select name, open_mode, database_role from v$database;

select process, status, sequence# from v$managed_standby;

ALTER DATABASE RECOVER  managed standby database  cancel;

shutdown immediate;

srvctl stop asm -f

Patch 1: on binary level

root user:

cd /users/oracle

. set_home_env +ASM

export PATH=$PATH:$GRID_HOME/OPatch

$GRID_HOME/OPatch/opatchauto apply $DOWNLOAD_PATCH/39062956/39036936

start asm

startup mount;

Patch 2: on Standby:

Stop DB - no need to stop ASM

remove alert log

From ORACLE_HOME

export PATH=$PATH:/usr/ccs/bin

$ORACLE_HOME/OPatch/opatch apply  $DOWNLOAD_PATCH/39062956/38906621

Starting logapply on standby:

alter database recover managed standby database disconnect from session;

On Primary:

alter system set log_archive_dest_state_2='ENABLE' scope=both;

Enjoy !

Your comments are most valuable and it would help me to come up with better posts. 

How to create swap file

 

How to Create, Verify, and Increase Swap Space on Linux

Swap space is disk space that Linux uses as virtual memory when physical RAM is exhausted. Proper swap configuration helps improve system stability and prevents out-of-memory situations.

PURPOSE: All documents are provided on this Blog just for educational purposes only.  Please make sure that you run it in your test environment before to move on to production environment. 

What is Swap Space?

Swap is a reserved area on disk that acts as an extension of RAM. When memory usage becomes high, inactive pages can be moved from RAM to swap, freeing memory for active processes.

Benefits of Swap

• Prevents application crashes due to memory shortages.
• Provides additional virtual memory.
• Supports system hibernation (when configured appropriately).
• Improves system stability during memory spikes.

Check Existing Swap Space

Use any of the following commands to verify current swap configuration and usage.

free -h

Check Active Swap Devices

swapon -s

cat /proc/swaps

swapon --show

Create a New Swap File

This method creates a 2 GB swap file. Adjust the size according to your requirements.

Step 1: Disable Existing Swap (if required)

swapoff -a
Step 2: Create Swap File
sudo fallocate -l 2G /swapfile
Verify file creation:
ls -lh /swapfile

Step 3: Set Correct Permissions
sudo chmod 600 /swapfile
Verify permissions:
ls -l /swapfile
Step 4: Format the File as Swap
sudo mkswap /swapfile

Step 5: Enable Swap
sudo swapon /swapfile

Step 6: Verify Swap Activation
swapon --show
or
free -h

Step 7: Make Swap Persistent After Reboot
Add the following entry to /etc/fstab:
echo "/swapfile none swap sw 0 0" | sudo tee -a /etc/fstab
Verify:
cat /etc/fstab
Expected entry:
/swapfile none swap sw 0 0

Increase Existing Swap Partition
Current swap size: 8GB
Target swap size: 16GB

Step 1: Disable Swap
swapoff -a
Verify:
swapon --show
No output should be displayed.
Step 2: Increase Virtual Disk Size in vCenter
Increate hard disk 1 from 64 to 90 GB
Navigate to VMware vCenter:
VM Settings
→ Hard Disk 1
→ Increase disk size
Example:
64 GB → 90 GB

Step 3: Detect New Disk Size
Check available disk space:
fdisk -l
If the new size is not visible, rescan the disk:
echo 1 > /sys/class/scsi_device/0\:0\:0\:0/device/rescan
Run again:
fdisk -l

Step 4: Modify Partitions
Launch partition manager:
cfdisk
Existing Layout
/dev/sda3  Linux filesystem
/dev/sda4  Linux swap
Delete Existing Swap Partition
Select:
/dev/sda4
Choose:
Delete
Resize Root/Data Partition 
cfdisk
Select:
/ dev/sda3  # This was a root partition which we increased on vCenter.
Choose:
Resize ==> Write

Create New Swap Partition
Use remaining free space:
New
Type → Linux swap

mkswap /dev/sda4
Setting up swapspace version 1, size = 26 GiB (27917262848 bytes)
no label, UUID=xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx
# edit the UUID
vi /etc/fstab

mount -a

swapon /dev/sda4
swapon --show
NAME      TYPE      SIZE USED PRIO
/dev/sda4 partition  26G   0B   -2

ORA-800_VKTM

 

ORA-800_VKTM

 

ORA-00800: soft external error, arguments: [Set Priority Failed], [VKTM]

ORA-800_VKTM_Fix

PURPOSE: All documents are provided on this Blog just for educational purposes only.  Please make sure that you run it in your test environment before to move on to production environment. 

After kernel update on linux system you get ORA-00800 error. In many cases you need to as follow

echo 950000 > /sys/fs/cgroup/cpu,cpuacct/user.slice/cpu.rt_runtime_us

The above command will fix the Issue, and you need to restart the DB. But when you reboot your Server the change will lost. Better to add following line in the rc.local file to persistent after reboot.

vi /etc/rc.local
echo 950000 > /sys/fs/cgroup/cpu,cpuacct/user.slice/cpu.rt_runtime_us

chmod +x /etc/rc.local

That's it 

I have script, if you have multiple server to fix the issue via jump host-.

#!/bin/bash

if [ -f $1 ]; then

 SERVERLIST=$(cat $1)

else

 SERVERLIST=$1

fi

for i in $SERVERLIST; do

    echo "Processing $i ..."

    ssh -q -T \

        -o ConnectTimeout=1 \

        -o PreferredAuthentications=publickey \

        "$i" /bin/bash --noprofile --norc <<'EOF'

echo 950000 > /sys/fs/cgroup/cpu,cpuacct/user.slice/cpu.rt_runtime_us

cat /sys/fs/cgroup/cpu,cpuacct/user.slice/cpu.rt_runtime_us

touch /etc/rc.local

grep -q "cpu.rt_runtime_us" /etc/rc.local || cat >> /etc/rc.local <<'EORC'

# Set CPU real-time runtime for Oracle VKTM

echo 950000 > /sys/fs/cgroup/cpu,cpuacct/user.slice/cpu.rt_runtime_us

EORC

chmod +x /etc/rc.local

EOF

done