Here I’m going to list all the errors/fixes during the an prechecks/upgrade to 26ai:
1:- Error during “gridSetup.sh -silent -upgrade -executePrereqs”
[oracle@rac-ol8-01 ~]$ /u01/app/26ai/grid/gridSetup.sh -silent -upgrade -executePrereqs
Launching Oracle Grid Infrastructure Setup Wizard...
*********************************************
Hard Limit: maximum open file descriptors: This is a prerequisite condition to test whether the hard limit for maximum open file descriptors is set correctly.
Severity: CRITICAL
Overall status: OPERATION_FAILED
-----------------------------------------------
-----------------------------------------------
Error message: PRVG-0341 : Failed to retrieve the current login shell from nodes "rac-ol8-01"
Cause: An attempt to retrieve the current login shell from indicated nodes failed.
Action: Ensure that the required login shell settings for current user are correct on the indicated nodes.
*********************************************
Soft Limit: maximum open file descriptors: This is a prerequisite condition to test whether the soft limit for maximum open file descriptors is set correctly.
Severity: CRITICAL
Overall status: OPERATION_FAILEDThis error is because process tries to ssh to itself and it needs password input, to avoid that, this is the fix:
[oracle@rac-ol8-01 ~]$ cd .ssh/
[oracle@rac-ol8-01 .ssh]$ cat id_rsa.pub >> authorized_keys
[oracle@rac-ol8-01 .ssh]$
[oracle@rac-ol8-01 .ssh]$ ssh rac-ol8-01
Activate the web console with: systemctl enable --now cockpit.socket
Last login: Fri Oct 2 16:42:46 2026
[oracle@rac-ol8-01 ~]$
** repeat the same on each rac node.2:- Error during “gridSetup.sh -silent -upgrade -executePrereqs”
[oracle@rac-ol8-01 ~]$ /u01/app/26ai/grid/gridSetup.sh -silent -upgrade -executePrereqs
Launching Oracle Grid Infrastructure Setup Wizard...
....
Error message: PRVH-1546 : Grid Infrastructure Management Repository database resource is configured in the source CRS home "/u01/app/oracle/product/19c/grid".
Cause: The Configuration Verification Utility (CVU) determined that the
Grid Infrastructure Management Repository (GIMR) database resource
configuration exists in the indicated source CRS home. Starting
with Oracle Database 23c, GIMR database is not supported.
Action: Ensure that the Grid Infrastructure Management Repository database
is deleted completely from the indicated source CRS home using the
commands ''srvctl stop mgmtdb -f; srvctl remove mgmtdb -f'' or run
the command ''<DEST_GI_HOME>/crs/install/reposScript.sh
-export_dir="location for export"'' from the destination CRS home
and retry the operation. Refer to My Oracle Support note 2972418.1
for more details.
Error message: Refer to My Oracle Support notes "2972418.1" for more details regarding errors "PRVH-1546".
Cause: Cause Of Problem Not Available
Action: User Action Not Available
[FATAL] [INS-13019] Some mandatory prerequisites are not met. These prerequisites cannot be ignored.
CAUSE: The following prerequisites are mandatory and cannot be ignored:
- Grid Infrastructure Management Repository cleanup.
ACTION: For detailed information of the failed prerequisites check the log: /u01/app/oraInventory/logs/GridSetupActions2026-10-02_04-51-40PM/gridSetupActions2026-10-02_04-51-40PM.log. Then either from the log file or from installation manual find the appropriate configuration to meet the prerequisites and fix it manually.
[oracle@rac-ol8-01 ~]$FIX (as error already suggest):
1:- Remove CHA dependencies:
[oracle@rac-ol8-01 ~]$ crsctl stat res ora.chad -p | grep mgmt
START_DEPENDENCIES=hard(global:ora.mgmtdb) pullup(global:ora.mgmtdb)
START_DEPENDENCIES_RTE_INTERNAL=<xml><Cond name="mgmtDBExist">False</Cond></xml>
START_DEPENDENCIES_TEMPLATE=<xml><If cond="mgmtDBExist" value="True"> <Then>hard(global:ora.mgmtdb) pullup(global:ora.mgmtdb)</Then> <Else></Else> </If></xml>
STOP_DEPENDENCIES=hard(global:intermediate:ora.mgmtdb)
STOP_DEPENDENCIES_RTE_INTERNAL=<xml><Cond name="mgmtDBExist">False</Cond></xml>
STOP_DEPENDENCIES_TEMPLATE=<xml><If cond="mgmtDBExist" value="True"><Then>hard(global:intermediate:ora.mgmtdb)</Then> <Else></Else> </If></xml>
[oracle@rac-ol8-01 ~]$
[oracle@rac-ol8-01 ~]$ crsctl modify resource ora.chad -attr "START_DEPENDENCIES='', STOP_DEPENDENCIES=''" -unsupported
[oracle@rac-ol8-01 ~]$ crsctl modify resource ora.chad -attr "START_DEPENDENCIES_RTE_INTERNAL='', STOP_DEPENDENCIES_RTE_INTERNAL=''" -unsupported
[oracle@rac-ol8-01 ~]$ crsctl modify resource ora.chad -attr "START_DEPENDENCIES_TEMPLATE='', STOP_DEPENDENCIES_TEMPLATE=''" -unsupported
[oracle@rac-ol8-01 ~]$ crsctl stat res ora.chad -p | grep mgmt
[oracle@rac-ol8-01 ~]$
** restart to make sure starts ok:
[oracle@rac-ol8-01 ~]$ srvctl stop cha
[oracle@rac-ol8-01 ~]$ srvctl start cha
[oracle@rac-ol8-01 ~]$ srvctl status cha
Oracle Cluster Health Analysis Service is enabled
Oracle Cluster Health Analysis Service is running on nodes: rac-ol8-02, rac-ol8-01
[oracle@rac-ol8-01 ~]$
2:- Delete MGMTDB DB
[oracle@rac-ol8-01 ~]$ srvctl status mgmtdb
Database is enabled
Instance -MGMTDB is running on node rac-ol8-01
[oracle@rac-ol8-01 ~]$
[oracle@rac-ol8-01 ~]$ dbca -silent -deleteDatabase -sourceDB -MGMTDB
3:- Remove listener:
[oracle@rac-ol8-01 ~]$ srvctl stop mgmtlsnr
[oracle@rac-ol8-01 ~]$ srvctl remove mgmtlsnr
[oracle@rac-ol8-01 ~]$3: Error during “cluvfyrac.sh stage -pre crsinst”
[oracle@rac-ol8-01 ~]$ /u01/app/26ai/grid/bin/cluvfyrac.sh stage -pre crsinst -n rac-ol8-01
....
Failures were encountered during execution of CVU verification request "stage -pre crsinst".
Domain Sockets ...FAILED
....
....
"rac-ol8-01".
rac-ol8-01: PRVG-11750 : File "/var/tmp/.oracle/ora_gipc_agent_ag_CSSDAG_z"
exists on node "rac-ol8-01".
rac-ol8-01: PRVG-11750 : File "/var/tmp/.oracle/ora_gipc_rac-ol8-01_GIPCD_lock"
exists on node "rac-ol8-01".
rac-ol8-01: PRVG-11750 : File "/var/tmp/.oracle/ora_gipc_rac-ol8-01_GPNPD_lock"
exists on node "rac-ol8-01".
rac-ol8-01: PRVG-11750 : File "/var/tmp/.oracle/ora_gipc_OHASD_soleowner_lock"
exists on node "rac-ol8-01".
rac-ol8-01: PRVG-11750 : File "/var/tmp/.oracle/mdnsd.pid" exists on node
"rac-ol8-01".
rac-ol8-01: PRVG-11750 : File "/var/tmp/.oracle/ora_gipc_rac-ol8-01_MOND"
exists on node "rac-ol8-01".
rac-ol8-01: PRVG-11750 : File "/var/tmp/.oracle/ora_gipc_GPNPD_rac-ol8-01_lock"
exists on node "rac-ol8-01".
rac-ol8-01: PRVG-11750 : File "/var/tmp/.oracle/ora_gipc_rac-ol8-01_MDNSD_lock"
exists on node "rac-ol8-01".
rac-ol8-01: PRVG-11750 : File "/var/tmp/.oracle/ora_gipc_rac-ol8-01_CHAD"
exists on node "rac-ol8-01".
rac-ol8-01: PRVG-11750 : File
"/var/tmp/.oracle/ora_gipc_scls_opct_rac-ol8-01_lock" exists on
node "rac-ol8-01".
rac-ol8-01: PRVG-11750 : File
"/var/tmp/.oracle/ora_gipc_monitor_ag_CSSDAG_z_lock" exists on node
"rac-ol8-01".
rac-ol8-01: PRVG-11750 : File "/var/tmp/.oracle/ora_gipc_rac-ol8-01_EVMD_lock"
exists on node "rac-ol8-01".
rac-ol8-01: PRVG-11750 : File "/var/tmp/.oracle/ora_gipc_rac-ol8-01_LOGD"
exists on node "rac-ol8-01".
rac-ol8-01: PRVG-11750 : File "/var/tmp/.oracle/ora_gipc_srac-ol8-01CRFM_SIPC"
exists on node "rac-ol8-01".
rac-ol8-01: PRVG-11750 : File
"/var/tmp/.oracle/ora_gipc_srac-ol8-01CRFM_CLIIPC_lock" exists on
node "rac-ol8-01".
rac-ol8-01: PRVG-11750 : File "/var/tmp/.oracle/ora_gipc_rac-ol8-01_LOGD_lock"
exists on node "rac-ol8-01".
rac-ol8-01: PRVG-11750 : File "/var/tmp/.oracle/ora_gipc_rac-ol8-01_GIPCD"
exists on node "rac-ol8-01".
rac-ol8-01: PRVG-11750 : File "/var/tmp/.oracle/mdnsd" exists on node
"rac-ol8-01".
rac-ol8-01: PRVG-11750 : File "/var/tmp/.oracle/ora_gipc_rac-ol8-01_CRSD_lock"
exists on node "rac-ol8-01".
rac-ol8-01: PRVG-11750 : File "/var/tmp/.oracle/ora_gipc_gipcd_rac-ol8-01"
exists on node "rac-ol8-01".
rac-ol8-01: PRVG-11750 : File
"/var/tmp/.oracle/ora_gipc_srac-ol8-01CRFM_MIIPC_lock" exists on
node "rac-ol8-01".
rac-ol8-01: PRVG-11750 : File "/var/tmp/.oracle/ora_gipc_rac-ol8-01_GPNPD"
exists on node "rac-ol8-01".
rac-ol8-01: PRVG-11750 : File "/var/tmp/.oracle/npohasd2" exists on node
"rac-ol8-01".
rac-ol8-01: PRVG-11750 : File "/var/tmp/.oracle/ora_gipc_rac-ol8-01_MOND_lock"
exists on node "rac-ol8-01".
rac-ol8-01: PRVG-11750 : File "/var/tmp/.oracle/ora_gipc_rac-ol8-01_MDNSD"
exists on node "rac-ol8-01".
rac-ol8-01: PRVG-11750 : File "/var/tmp/.oracle/ora_gipc_GPNPD_rac-ol8-01"
exists on node "rac-ol8-01".
Grid Infrastructure home path: /u01/app/oracle/product/19c/grid ...FAILED
'/u01/app/oracle/product/19c/grid' ...FAILED
rac-ol8-01: PRVG-11931 : Path "/u01/app/oracle/product/19c/grid" is not
writeable on node "rac-ol8-01".
RPM Package Manager database ...INFORMATION
PRVG-11250 : The check "RPM Package Manager database" was not performed because
it needs 'root' user privileges.
Refer to My Oracle Support notes "2548970.1" for more details regarding errors
PRVG-11250".
ORAchk checks ...WARNING
Verify rngd system service ...WARNING
rac-ol8-01: AHF-4993: system service rngd is not running
Additional Error Details:
DATA FROM RAC-OL8-01 - VERIFY RNGD SYSTEM SERVICE
Redirecting to /bin/systemctl status rngd.service
Unit rngd.service could not be found.
CVU operation performed: stage -pre crsinst
Date: 02-Oct-2026 16:42:51
CVU version: 19.32.0.0.0 (070326x8664)
Clusterware version: 19.0.0.0.0
CVU home: /u01/app/oracle/product/19c/grid
Grid home: /u01/app/oracle/product/19c/grid
User: oracle
Operating system: Linux5.15.0-324.217.5.2.el8uek.x86_64
[oracle@rac-ol8-01 ~]$Well… this is not an error. We are upgrading on a server currently running with the 19c version, so of course you have LOCK files and ORACLE_HOME locked (not writeable)
4:- rootupgrade.sh FAILS
[root@rac-ol8-01 ~]# /u01/app/26ai/grid/rootupgrade.sh
Check /u01/app/26ai/grid/install/root_rac-ol8-01_2026-10-02_21-56-07-829957486.log for the output of root script
[root@rac-ol8-01 ~]#
[oracle@rac-ol8-01 ~]$ tail -f /u01/app/26ai/grid/install/root_rac-ol8-01_2026-10-02_21-56-07-829957486.log
.....
2026/10/02 22:11:36 CLSRSC-595: Executing upgrade step 4 of 16: 'UpgPrechecks'.
2026/10/02 22:12:20 CLSRSC-595: Executing upgrade step 5 of 16: 'SetupOSD'.
2026/10/02 22:12:21 CLSRSC-595: Executing upgrade step 6 of 16: 'PreUpgrade'.
2026/10/02 22:12:29 CLSRSC-486:
At this stage of upgrade, the OCR has changed.
Any attempt to downgrade the cluster after this point will require a complete cluster outage to restore the OCR.
2026/10/02 22:12:29 CLSRSC-541:
To downgrade the cluster:
1. All nodes that have been upgraded must be downgraded.
2026/10/02 22:12:30 CLSRSC-542:
2. Before downgrading the last node, the Grid Infrastructure stack on all other cluster nodes must be down.
2026/10/02 22:12:34 CLSRSC-468: Setting Oracle Clusterware and ASM to rolling migration mode
2026/10/02 22:12:34 CLSRSC-482: Running command: '/u01/app/oracle/product/19c/grid/bin/crsctl start rollingupgrade 23.0.0.0.0'
CRS-1131: The cluster was successfully set to rolling upgrade mode.
2026/10/02 22:12:43 CLSRSC-482: Running command: '/u01/app/26ai/grid/bin/asmca -silent -upgradeNodeASM -nonRolling false -oldCRSHome /u01/app/oracle/product/19c/grid -oldCRSVersion 19.0.0.0.0 -firstNode true -startRolling false '
2026/10/02 22:12:51 CLSRSC-469: Successfully set Oracle Clusterware and ASM to rolling migration mode
2026/10/02 22:12:54 CLSRSC-1009: failed to stop resource OC4J
Died at /u01/app/26ai/grid/crs/install/oraqosmserver.pm line 304.There is no OC4J:
[root@rac-ol8-02 ~]# . oraenv
ORACLE_SID = [root] ? +ASM2
The Oracle base has been set to /u01/app/grid
[root@rac-ol8-02 ~]# crsctl stat res ora.oc4j -p
CRS-2613: Could not find resource 'ora.oc4j'.
[root@rac-ol8-02 ~]# srvctl config oc4j
Exception in thread "main" java.lang.RuntimeException: unknown noun 112
at oracle.ops.opsctl.Action.execute(Action.java:651)
at oracle.ops.opsctl.OPSCTLDriver.execute(OPSCTLDriver.java:572)
at oracle.ops.opsctl.OPSCTLDriver.main(OPSCTLDriver.java:291)
[root@rac-ol8-02 ~]# crsctl stat res ora.qosmserver -p
CRS-2613: Could not find resource 'ora.qosmserver'.
[root@rac-ol8-02 ~]# crsctl stat res ora.oc4j -p
CRS-2613: Could not find resource 'ora.oc4j'.
[root@rac-ol8-02 ~]#For the Fix, during gridSetup.sh the apply the latest RU to the new 26ai grid home (JUL 26 on this case):
[oracle@rac-ol8-01 grid]$ ./gridSetup.sh -silent -applyRU /u01/installs/26ai/RU/39568945
[INFO] [INS-32830] Options -applyRU, -applyPSU and -applyRUR are deprecated in this release. For more details, refer to help documentation.
Two operations will be performed: first the home will be patched, and then the installer will be launched.
Operation 1:
Preparing the home to patch...
Applying the patch /u01/installs/26ai/RU/39568945...
Successfully applied the patch.
The log can be found at: /u01/app/oraInventory/logs/GridSetupActions2026-10-05_09-12-49AM/installerPatchActions_2026-10-05_09-12-49AM.log
Operation 2:
Launching Oracle Grid Infrastructure Setup Wizard...
[FATAL] [INS-40426] Grid installation option has not been specified.
ACTION: Specify the valid installation option.
[oracle@rac-ol8-01 grid]$
Comments