OpenSolaris

Printable Version Enter a New Search
Bug ID 6531649
Synopsis oracle wizard failing because it does not create lh when called for the 1st time on a node
State 10-Fix Delivered (Fix available in build)
Category:Subcategory suncluster:ds-config
Keywords
Responsible Engineer Ritu Agrawal
Reported Against
Duplicate Of
Introduced In
Commit to Fix 3.2_patch_04
Fixed In 3.2_patch_04
Release Fixed 3.2_patch(3.2_patch_04)
Related Bugs
Submit Date 6-March-2007
Last Update Date 30-March-2007
Description
Running:
 
 sc32_GA + fix for 6513885
 sap & ha-oracle agent pkgs from sc32ds_patch_b2
 s10u3 SPARC 

Problem:

On runs of oracle wizard where I ask it to create a new lh and hasp rs, if the run is the first run after a bootup of the node, the wizard fails to create the lh and subsequently the ora lsnr fails to start causing the wizard to fail.  In reproducing this problem with Ritu we found that if the user then re-runs the oracle wizard the same way in the same node the wizard works correctly.  As of now it appears that this problem is seen only when wizard is run for the first time on the node after its booted.  Some logs follow.

Logs showing failure:
~~~~~~~~~~~~~~~~~~~~~

***Wizard window showing list of objects to be created -

    The following Sun Cluster configuration will be created.

    To view the details for an option, select the option.

                   Name                       Value
                   ====                       =====
        1) Server Resource Name     oradb
        2) Listener Resource Name   oralsnr
        3) Resource Group Name      oracle-rg
        4) Node List                probin1,probin3,probin4
        5) Highly Available Stora...hasp4ora
        6) File-System Mount Points
/oracle/stage/817_64,/oracle/COK/origlogA...
        7) Logical Hostname Resou...robin-3
        8) Logical Hostname         robin-3
        9) IPMP Group List         
sc_ipmp0@probin1,sc_ipmp0@probin3,sc_ipmp...

   >>> View Results  <<<

***Wizard failure -

             ERROR: The following error occurred during the creation of the
             configuration
             
             oralsnr: resource group failed to start on chosen node; it
may end
             up failing over to other node(s)



             ERROR: Non-zero exit status

    The following Sun Cluster commands were run
    
    Configuration commands...


    /usr/cluster/bin/clresourcegroup create -p
nodelist=probin1,probin3,probin4
    oracle-rg

    /usr/cluster/bin/clresource create -t SUNW.HAStoragePlus:4 -g
oracle-rg -p
    FilesystemMountPoints=/oracle/stage/817_64,/oracle/COK/origlogA,
    /oracle/COK/origlogB,/oracle/COK/mirrlogA,/oracle/COK/mirrlogB,
    /oracle/COK/saparch,/oracle/COK/sapreorg,/oracle/COK/sapdata1,
    /oracle/COK/sapdata2,/oracle/COK/sapdata3,/oracle/COK/sapdata4,
    /oracle/COK/sapdata5,/oracle/COK/sapdata6,/oracle/client/81x_64,
    /oracle/COK/saptrace,/oracle/COK/920_64,/oracle/client/92x_64 hasp4ora

    /usr/cluster/bin/clresourcegroup online -emM oracle-rg

    /usr/cluster/bin/clresource create -t SUNW.oracle_server:6 -g
oracle-rg -p
    Resource_dependencies=hasp4ora -p ORACLE_HOME=/oracle/COK/920_64 -p
    ORACLE_SID=COK -p
    Custom_action_file=/oracle/COK/920_64/sc_stuff/scor_action -p
    Parameter_file=/oracle/COK/920_64/dbs/initCOK.ora -p
    Alert_log_file=/oracle/COK/920_64/rdbms/log/alert_COK.log -p
...

***From commandlog -

03/02/2007 12:34:24 probin1 19163 root START - clresourcetype "register"
"SUNW.oracle_server"
03/02/2007 12:34:24 probin1 19163 root END 0
03/02/2007 12:34:39 probin1 19165 root START - clresourcetype "register"
"SUNW.oracle_listener"
03/02/2007 12:34:39 probin1 19165 root END 0
03/02/2007 12:51:11 probin1 19341 root START - clresourcegroup "create"
-p "nodelist=probin1,probin3,probin4" "oracle-rg"
03/02/2007 12:51:11 probin1 19341 root END 0
03/02/2007 12:51:14 probin1 19343 root START - clresource "create" -t
"SUNW.HAStoragePlus:4" -g "oracle-rg" -p
"FilesystemMountPoints=/oracle/stage/817_64,/oracle/COK/origlogA,/oracle/COK/origlogB,/oracle/COK/mirrlogA,/oracle/COK/mirrlogB,/oracle/COK/saparch,/oracle/COK/sapreorg,/oracle/COK/sapdata1,/oracle/COK/sapdata2,/oracle/COK/sapdata3,/oracle/COK/sapdata4,/oracle/COK/sapdata5,/oracle/COK/sapdata6,/oracle/client/81x_64,/oracle/COK/saptrace,/oracle/COK/920_64,/oracle/client/92x_64"
"hasp4ora"
03/02/2007 12:51:16 probin1 19343 root END 0
03/02/2007 12:51:16 probin1 19348 root START - clresourcegroup "online"
"-emM" "oracle-rg"
03/02/2007 12:51:18 probin1 19348 root END 0
03/02/2007 12:51:20 probin1 19362 root START - clresource "create" -t
"SUNW.oracle_server:6" -g "oracle-rg" -p
"Resource_dependencies=hasp4ora" -p "ORACLE_HOME=/oracle/COK/920_64" -p
"ORACLE_SID=COK" -p
"Custom_action_file=/oracle/COK/920_64/sc_stuff/scor_action" -p
"Parameter_file=/oracle/COK/920_64/dbs/initCOK.ora" -p
"Alert_log_file=/oracle/COK/920_64/rdbms/log/alert_COK.log" -p
"Connect_string=scott/tiger" "oradb"
03/02/2007 12:51:38 probin1 19362 root END 0
03/02/2007 12:51:38 probin1 19616 root START - clresourcegroup "online"
"-emM" "oracle-rg"
03/02/2007 12:51:38 probin1 19616 root END 0
03/02/2007 12:51:41 probin1 19626 root START - clresource "create" -t
"SUNW.oracle_listener:5" -g "oracle-rg" -p
"Resource_dependencies=hasp4ora" -p "ORACLE_HOME=/oracle/COK/920_64"
"oralsnr"
03/02/2007 12:51:47 probin1 19626 root END 0
03/02/2007 12:51:48 probin1 19715 root START - clresource "disable"
"hasp4ora"
03/02/2007 12:51:49 probin1 19715 root END 0
03/02/2007 12:51:53 probin1 19726 root START - clresource "delete"
"hasp4ora"
03/02/2007 12:51:54 probin1 19726 root END 0

Logs showing success:
~~~~~~~~~~~~~~~~~~~~~

***wizard log fragment -

   >>> View Results  <<<

    The following Sun Cluster commands were run 
    
    Configuration commands...


    /usr/cluster/bin/clresourcegroup create -p nodelist=probin4,probin3,probin1
    oracle-rg

    /usr/cluster/bin/clresource create -t SUNW.HAStoragePlus:4 -g oracle-rg -p
    FilesystemMountPoints=/oracle/stage/817_64,/oracle/COK/origlogA,
    /oracle/COK/origlogB,/oracle/COK/mirrlogA,/oracle/COK/mirrlogB,
    /oracle/COK/saparch,/oracle/COK/sapreorg,/oracle/COK/sapdata1,
    /oracle/COK/sapdata2,/oracle/COK/sapdata3,/oracle/COK/sapdata4,
    /oracle/COK/sapdata5,/oracle/COK/sapdata6,/oracle/client/81x_64,
    /oracle/COK/saptrace,/oracle/COK/920_64,/oracle/client/92x_64 hasp4ora

    /usr/cluster/bin/clresourcegroup online -emM oracle-rg

    /usr/cluster/bin/clreslogicalhostname create -g oracle-rg -h robin-3 -N
    sc_ipmp0@probin4,sc_ipmp0@probin3,sc_ipmp0@probin1 robin-3

    /usr/cluster/bin/clresourcegroup online -emM oracle-rg

    /usr/cluster/bin/clresource create -t SUNW.oracle_server:6 -g oracle-rg -p
    Resource_dependencies=hasp4ora -p ORACLE_HOME=/oracle/COK/920_64 -p
    ORACLE_SID=COK -p Parameter_file=/oracle/COK/920_64/dbs/initCOK.ora -p
    Alert_log_file=/oracle/COK/saptrace/background/alert_COK.log -p
    Connect_string=scott/tiger oradb

    /usr/cluster/bin/clresourcegroup online -emM oracle-rg

    /usr/cluster/bin/clresource create -t SUNW.oracle_listener:5 -g oracle-rg -
    p Resource_dependencies=hasp4ora -p ORACLE_HOME=/oracle/COK/920_64 oralsnr

    /usr/cluster/bin/clresourcegroup online -emM oracle-rg


    Configuration of Sun Cluster HA for Oracle succeeded


    Press RETURN to close the wizard

***Corresponding commandlog msgs -

03/06/2007 12:22:02 probin4 8293 root START - clresourcegroup "create" -p "nodelist=probin4,probin3,probin1" "oracle-rg"
03/06/2007 12:22:03 probin4 8293 root END 0
03/06/2007 12:22:06 probin4 8295 root START - clresource "create" -t "SUNW.HAStoragePlus:4" -g "oracle-rg" -p "FilesystemMountPoints=/oracle/stage/817_64,/oracle/COK/origlogA,/oracle/COK/origlogB,/oracle/COK/mirrlogA,/oracle/COK/mirrlogB,/oracle/COK/saparch,/oracle/COK/sapreorg,/oracle/COK/sapdata1,/oracle/COK/sapdata2,/oracle/COK/sapdata3,/oracle/COK/sapdata4,/oracle/COK/sapdata5,/oracle/COK/sapdata6,/oracle/client/81x_64,/oracle/COK/saptrace,/oracle/COK/920_64,/oracle/client/92x_64" "hasp4ora"
03/06/2007 12:22:08 probin4 8295 root END 0
03/06/2007 12:22:08 probin4 8306 root START - clresourcegroup "online" "-emM" "oracle-rg"
03/06/2007 12:22:10 probin4 8306 root END 0
03/06/2007 12:22:13 probin4 8320 root START - clreslogicalhostname "create" -g "oracle-rg" -h "robin-3" -N "sc_ipmp0@probin4,sc_ipmp0@probin3,sc_ipmp0@probin1" "robin-3"
03/06/2007 12:22:16 probin4 8320 root END 0
03/06/2007 12:22:16 probin4 8336 root START - clresourcegroup "online" "-emM" "oracle-rg"
03/06/2007 12:22:16 probin4 8336 root END 0
03/06/2007 12:22:19 probin4 8338 root START - clresource "create" -t "SUNW.oracle_server:6" -g "oracle-rg" -p "Resource_dependencies=hasp4ora" -p "ORACLE_HOME=/oracle/COK/920_64" -p "ORACLE_SID=COK" -p "Parameter_file=/oracle/COK/920_64/dbs/initCOK.ora" -p "Alert_log_file=/oracle/COK/saptrace/background/alert_COK.log" -p "Connect_string=scott/tiger" "oradb"
03/06/2007 12:22:32 probin4 8338 root END 0
03/06/2007 12:22:33 probin4 8592 root START - clresourcegroup "online" "-emM" "oracle-rg"
03/06/2007 12:22:33 probin4 8592 root END 0
03/06/2007 12:22:37 probin4 8602 root START - clresource "create" -t "SUNW.oracle_listener:5" -g "oracle-rg" -p "Resource_dependencies=hasp4ora" -p "ORACLE_HOME=/oracle/COK/920_64" "oralsnr"
03/06/2007 12:22:43 probin4 8602 root END 0
03/06/2007 12:22:44 probin4 8770 root START - clresourcegroup "online" "-emM" "oracle-rg"
03/06/2007 12:22:44 probin4 8770 root END 0
I just reproduced this problem on a different s10u3 sparc cluster running core and ha-oracle bits from sc32_patch_b3.
Work Around
N/A
Comments
N/A