Blogs

ODA Pre-check Report Alert (Cluster Ready Services Attribute STOP_DEPENDENCIES/START_DEPENDENCIES for Database is Incorrect)

I am planning to Patch Oracle Database Appliance (ODA) from version 19.11 to 19.15 on X8-2L.

I am not discussing the whole ODA patching here but one of the issues during the pre-patch report.

When I ran the pre-patch report I got the below error/alert… This is part of the pre-check report.

__GI__
Validate GI metadata            Success   Successfully validated GI metadata
Validate supported GI versions  Success   Validated minimum supported versions.
Validate available space        Success   Validated free space under /u01
Is clusterware running          Success   Clusterware is running
Validate patching tag           Success   Validated patching tag: 19.15.0.0.0.
Is system provisioned           Success   Verified system is provisioned
Validate ASM in online          Success   ASM is online
Validate kernel log level       Success   Successfully validated the OS log
                                          level
Validate minimum agent version  Success   GI patching enabled in current
                                          DCSAGENT version
Validate Central Inventory      Success   oraInventory validation passed
Validate patching locks         Success   Validated patching locks
Validate clones location exist  Success   Validated clones location
Validate DB start dependencies  Alert     Cluster Ready Services attribute
                                          START_DEPENDENCIES for database
                                          [testoda] is incorrect and will be
                                          corrected as part of server patching.
Validate DB stop dependencies   Alert     Cluster Ready Services attribute
                                          STOP_DEPENDENCIES for database
                                          [testoda] is incorrect and will be
                                          corrected as part of server patching.
Evaluate GI patching            Success   Successfully validated GI patching
Validate command execution      Success   Validated command execution                                          
                                          

 

Steps to Resolve:

1. First I check the DCS logs to see why it’s alerting on the CRS attribute is incorrect for the tested database.

run: cmd= '[/u01/app/19.11.0.0/grid/bin/crsctl,
 status,
 resource,
 -w,
 (TYPE = ora.database.type) AND (START_DEPENDENCIES coi 'orahome_sh') AND ((START_DEPENDENCIES nci 'odabase_n0') OR (START_DEPENDENCIES nci 'odabase_n1')),
 -v,
 -attr,
 NAME]'
2023-10-18 10:47:03,706 DEBUG [Thread-981] [] c.o.d.c.u.CommonsUtils: Output :
NAME=ora.testoda.db
2023-10-18 10:47:03,706 DEBUG [Thread-981] [] c.o.d.c.u.CommonsUtils: Output : EMPTY CONTENT
2023-10-18 10:47:03,712 DEBUG [Run patching pre-checks for component: GI : JobId=e13279b9-990a-4c45-b0f5-1a96740828e5] [] c.o.d.a.u.OraHomeStorageUtils: Processing string "NAME=ora.testoda.db"
2023-10-18 10:47:03,712 INFO [Run patching pre-checks for component: GI : JobId=e13279b9-990a-4c45-b0f5-1a96740828e5] [] c.o.d.a.u.OraHomeStorageUtils: START_DEPENDENCIES invalid for "testoda"

If we see it mention 3 start dependencies in logs.

 

2. Then I thought let me check what are the CRS start and stop dependencies attribute values for this database with issue and another database without issue.

Attribute of DB with Issue:

[grid@odanode ~]$ crsctl status resource ora.testoda.db -f | grep DEPENDENCIES=
START_DEPENDENCIES=hard(global:uniform:ora.RECO.dg, global:uniform:ora.DATA.dg, ora.data.orahome_sh.acfs) pullup(ora.data.orahome_sh.acfs, global:ora.DATA.dg, global:ora.RECO.dg) weak(type:ora.listener.type,global:type:ora.scan_listener.type,uniform:ora.ons,global:ora.gns)
STOP_DEPENDENCIES=hard(global:intermediate:ora.asm,global:shutdown:ora.RECO.dg,global:shutdown:ora.DATA.dg,ora.data.orahome_sh.acfs)
[grid@odanode ~]$

Attribute of DB without Issue:

[grid@odanode ~]$ crsctl status resource ora.devoda.db -f | grep DEPENDENCIES=
START_DEPENDENCIES=hard(global:uniform:ora.DATA.dg, ora.data.orahome_sh.acfs, ora.data.odabase_n0.acfs, ora.data.odabase_n1.acfs, global:uniform:ora.RECO.dg) pullup(ora.data.orahome_sh.acfs, ora.data.odabase_n0.acfs, ora.data.odabase_n1.acfs, global:ora.RECO.dg, global:ora.DATA.dg) weak(type:ora.listener.type,global:type:ora.scan_listener.type,uniform:ora.ons,global:ora.gns)
STOP_DEPENDENCIES=hard(global:intermediate:ora.asm,global:shutdown:ora.DATA.dg,ora.data.orahome_sh.acfs,intermediate:ora.data.odabase_n0.acfs,intermediate:ora.data.odabase_n1.acfs, global:shutdown:ora.RECO.dg)
[grid@odanode ~]$

 

3. Then I compared the working and nonworking attributes from above and saw below the missing nonworking attribute, which somewhat matches the logs that I show in Step 1.

missing from stop_dependencies = intermediate:ora.data.odabase_n0.acfs,intermediate:ora.data.odabase_n1.acfs

missing from start_dependencies = hard(ora.data.odabase_n0.acfs, ora.data.odabase_n1.acfs) pullup(ora.data.odabase_n0.acfs, ora.data.odabase_n1.acfs)

 

4. So, I modified the incorrect start and stop dependencies for the database shown in the pre-patch report using the below command. Basically, I copy the attribute from the correct DB.

[oracle@odanode ~] /u01/app/19.11.0.0/grid/bin/crsctl modify resource ora.testoda.db -attr "START_DEPENDENCIES='hard(global:uniform:ora.DATA.dg, ora.data.orahome_sh.acfs, ora.data.odabase_n0.acfs, ora.data.odabase_n1.acfs, global:uniform:ora.RECO.dg) pullup(ora.data.orahome_sh.acfs, ora.data.odabase_n0.acfs, ora.data.odabase_n1.acfs, global:ora.RECO.dg, global:ora.DATA.dg) weak(type:ora.listener.type,global:type:ora.scan_listener.type,uniform:ora.ons,global:ora.gns)'" -unsupported

[oracle@odanode ~] /u01/app/19.11.0.0/grid/bin/crsctl modify resource ora.testoda.db -attr "STOP_DEPENDENCIES=hard(global:intermediate:ora.asm,global:shutdown:ora.DATA.dg,ora.data.orahome_sh.acfs,intermediate:ora.data.odabase_n0.acfs,intermediate:ora.data.odabase_n1.acfs, global:shutdown:ora.RECO.dg)'" -unsupported

NOTE: If I don't use the last "-unsupported" keyword it gives below error.

CRS-4995:  The command 'Modify  resource' is invalid in crsctl. Use srvctl for this command.

 

5. After modifying the attribute, I ran the pre-patch report and didn’t see the alert anymore.

More from this Author