How to Configure RHP Server for Oracle FPP Local Mode in 19c Oracle RAC

INTRO: In 19c environments, enabling FPP local mode requires a Grid Infrastructure Management Repository (GIMR)

The Oracle Fleet Patching and Provisioning (FPP) local mode, introduced in 18c, enables local automation for many lifecycle operations. In Oracle FPP 23ai, this capability has been further simplified and streamlined.

However, in 19c environments, configuring FPP local mode requires a Grid Infrastructure Management Repository (GIMR) on the cluster that is being patched. In environments where GIMR-related components (such as mgmtdb, mgmtlsnr, and Cluster Health Monitor) have been removed, or where Grid Infrastructure was originally installed without them (since it is optional), these resources must be added back to use rhpserver locally.

You may follow Ludovico Caldara’s blog post: “Oracle Grid Infrastructure 19c does not configure FPP in local-mode by default. How to add it?” to perform the configuration manually. However, there is a simpler approach to add the required GIMR components. I will show it on a 19.30 Oracle RAC Database.

From Oracle Support, you can download the MDBUtil tool (Doc ID 2065175.1) – GI Management Repository configuration tool.

You can then run the following command to recreate the required repository database(GIMR) and specify the storage location:

[oracle@blt01 ~]$ ./mdbutil.pl --addmdb --target=+VOTE
mdbutil.pl version : 1.102
2026-05-20 10:53:10: I Starting To Configure MGMTDB at +VOTE...
2026-05-20 10:53:18: I Container database creation in progress... for GI 19.0.0.0.0
2026-05-20 11:19:03: I Plugable database creation in progress...
2026-05-20 11:22:22: I Executing "/tmp/mdbutil.pl --addchm" on blt01 as root to configure CHM.
root@blt01's password:
2026-05-20 11:26:41: W Not able to execute "/tmp/mdbutil.pl --addchm" on blt01 as root to configure CHM.
2026-05-20 11:26:42: I Executing "/tmp/mdbutil.pl --addchm" on blt02 as root to configure CHM.
root@blt02's password:
2026-05-20 11:26:48: W Not able to execute "/tmp/mdbutil.pl --addchm" on blt02 as root to configure CHM.
2026-05-20 11:26:48: I MGMTDB & CHM configuration done!

I have taken a warning about CHM resource (ora.crf).

[oracle@blt01 ~]$ ./mdbutil.pl --status
mdbutil.pl version : 1.102
2026-05-20 13:00:09: I Checking CHM status...
2026-05-20 13:00:11: I Listener MGMTLSNR is configured and running on blt02
2026-05-20 13:00:15: I Database MGMTDB is configured and running on blt02
2026-05-20 13:00:15: W Cluster Health Monitor (CHM) is configured and not running on blt01!

Adding CHM (ora.crf) resource manually

We will run the commands provided below on all RAC database nodes.

[root@blt01 ~]# crsctl modify res ora.crf -attr ENABLED=1 -init
[root@blt01 ~]# crsctl start res ora.crf -init
CRS-2672: Attempting to start 'ora.crf' on 'blt01'
CRS-2676: Start of 'ora.crf' on 'blt01' succeeded
[root@blt02 ~]# crsctl modify res ora.crf -attr ENABLED=1 -init
[root@blt02 ~]# crsctl start res ora.crf -init
CRS-2672: Attempting to start 'ora.crf' on 'blt02'
CRS-2676: Start of 'ora.crf' on 'blt02' succeeded
[oracle@blt01 ~]$  ./mdbutil.pl --addchm 
[oracle@blt02 ~]$  ./mdbutil.pl --addchm 
[oracle@blt02 ~]$ ./mdbutil.pl --status
mdbutil.pl version : 1.102
2026-05-20 19:17:44: I Checking CHM status...
2026-05-20 19:17:55: I Listener MGMTLSNR is configured and running on blt02
2026-05-20 19:18:15: I Database MGMTDB is configured and running on blt02
2026-05-20 19:18:18: I Cluster Health Monitor (CHM) is configured and running
--------------------------------------------------------------------------------

Use the srvctl binary located in the Grid Infrastructure Home (GRID_HOME). By default, the rhpserver configuration is as follows:

[oracle@blt01 ~]$ /u01/app/19.30/grid/bin/srvctl config rhpserver
Storage base path: /opt/oracle/rhp_images
Disk Groups: +VOTE
Port number: 8896
Temporary Location:
Transfer port range:
Rapid Home Provisioning Server is enabled
Rapid Home Provisioning Server is individually enabled on nodes:
Rapid Home Provisioning Server is individually disabled on nodes:
Email address:
Mail server address:
Mail server port:
Transport Level Security disabled
HTTP Secure is enabled
Endpoint : gns

I will use a dedicated local disk to store software images(database and grid) , not an asm diskgroup. Since there is no command to modify storage option of rhpserver, I will remove the rhpserver and add it back with root user.

[root@blt01 swrepo]#  /u01/app/19.30/grid/bin/srvctl remove rhpserver
[root@blt01 swrepo]#  /u01/app/19.30/grid/bin/srvctl add rhpserver -local -storage /swrepo
[oracle@blt01 ~]$ /u01/app/19.30/grid/bin/srvctl config rhpserver
Storage base path: /swrepo
Disk Groups:
Port number: 8896
Temporary Location:
Transfer port range:
Rapid Home Provisioning Server is enabled
Rapid Home Provisioning Server is individually enabled on nodes:
Rapid Home Provisioning Server is individually disabled on nodes:
Email address:
Mail server address:
Mail server port:
Transport Level Security disabled
HTTP Secure is enabled
Endpoint : gns

Our local RHP server is now ready for use. We can use it for Fleet Patching and Provisioning, as well as for Oracle Update Advisor to assess our Grid and database environment patch levels. For more information about Oracle Update Advisor, please refer to my blog post.

Hope it helps.