Patch-ID# 145335-22

22195397 oracle_listener:5 fails to probe the listener when configured w/ TLS
23228192 Solaris Cluster HA-Oracle with Data Guard starts DB incorrectly
 
(from 145335-21)
 
22151224 SUNW.scalable_asm_instance_proxy: "hatimerun: Error: -t timeOutSecs must be positive integer."
 
(from 145335-20)
 
19564893 clsetup for RAC RS fails on sun.storage_proxy.type dependency update validation
 
(from 145335-19)
 
18750307 abort_oracle function kills process belonging to different SID
 
(from 145335-18)
 
16581850 ASM diskgroup proxy goes offline during Oracle binaries installation
 
(from 145335-17)
 
15820974 script asmdg_proxy_daemon needs to use the v$asm_diskgroup_stat be used for the probe not v$asm_disk
16582483 10gR2 DB instance may not start due to ASM agent incorrect status
 
(from 145335-16)
 
16347826 need to add all changed binaries to U2 patches
 
(from 145335-15)
 
15808435 RAC proxy resource creation fails when cluster nodes hostname has upper-case letters
15808863 wizard won't discover the ASM sid
15810365 HA-Oracle requires changes to work with Oracle Solaris Cluster
15814020 RAC wizard failing w/ "ERROR: Oracle ASM is either not installed or the installation is invalid!"
15822430 HA-Oracle/ACFS fails with Oracle Clusterware ACFS resource not found for ACFS file system
15826127 SUNW.scalable_rac_server_proxy needs to tolerate a node name that uses a fully-qualified node name
15865889 wizard doesn't discover existing ASM and unable to create ASM resources
 
(from 145335-14)
 
        This revision accumulates feature point patch 149491-03 into OSC 3.3 U2.
 
(from 145335-13)
 
7180387 fix for 7057373 was not correctly implemented in Cluster 3.3 and above
 
(from 145335-12)
 
6478836 RAC 9i server and monitor core dump while CVM devices are disabled
6824735 HA-Oracle oracle_server_start times out on warning ORA-28002 message
7109516 user credentials for HA Oracle listener should be derived from tnslsnr and not from oracle-server
7113369 HA-Oracle fails to start listener if LISTENER_PATH_LENGTH is 79 bytes
 
(from 145335-11)
 
7151252 one database instance not starting after SUNW.scalable_acfs_proxy RS is disabled then re-enabled
 
(from 145335-10)
 
7009633 support RAC running in zone cluster with DBhome on ACFS
7087730 HA-Oracle server and listener probes cause high load on a NFS filer
7093487 "# rows selected" is printed in the list of ASM DGs found by the wizard
 
(from 145335-09)
 
7048227 HA-Oracle fails on Solaris 11 cluster
7048283 /opt/SUNWscor/oracle_server/bin/oracle_server_validate[1712]: syntax error at line 1752 : `)' unexpected
7057373 oracle_server_start fails when ORA-10873 returned from Oracle database
7057691 RAC /HA-Oracle wizards failing saying that ASM installation is invalid though manually OK
7058114 scalable_asm_instance_proxy rs going to stop failed state every time it is disabled
7061263 unnecessary msg about "srvctl start dg -g <dg> -n <node>" failing when resource is already running
7070523 continuous su: 'su root' succeeded for root on /dev/??? when Oracle 11.2.0.3 is installed
7082865 "Failed to update the INCARNATION value within the CCR TABLE GIConfig_ora-db2-rs"
 
(from 145335-08)
 
7044445 errors on the console when unmanage/manage an RG containing a SUNW.scalable_acfs_proxy resource
 
(from 145335-07)
 
7022615 SUNW.scalable_acfs_proxy RS in zone cluster must check affinity on RG for SUNW.wait_zc_boot RS
7026529 "oracle_server_validate" fails if ACLs are set on $ORACLE_HOME/bin/oracle
7029829 OSC agent for Oracle DB needs to avert race conditions and clear text password
7030399 problem with Oracle Solaris Cluster HA-OPS
7030417 problem with Oracle Solaris Cluster HA-OPS
7036088 Oracle RAC 11gR2 DB instances / rac_server_proxy RS may not start upon zone cluster boot up
7040284 HA-Oracle pre-11gR2 single instance ASM fails to start ASM - Oracle CSS is not running
7040833 unwanted ASM messages on zone cluster console following Live Upgrade
7042000 /opt/SUNWscor/oracle_server/bin/oracle_server_validate[33]: bin/ls: not found
7042935 unable to start SUNW.scalable_asm_instance_proxy and SUNW.scalable_asm_diskgroup_proxy resources
7043985 RAC data service validate methods will fail setuid bit check on certain NAS devices
7044568 Oracle ASM ORACLE_SID (+ASM1) does not match the Oracle ASM configuration ORACLE_SID () within CRS
 
(from 145335-06)
 
7028555 OSC agent for Oracle ASM needs to avert race conditions
7028616 OSC agent for Oracle ASM needs to use explicit pathnames when executing commands
7030267 OSC agent for ASM related RTs should not execute non-root application commands as root
7030284 OSC agent for Oracle DB RT should not execute non-root application commands as root
 
(from 145335-05)
 
7035295 need to add all changed binaries to U1 patches
 
(from 145335-04)
 
7025260 HA-Oracle DB primary node reboot caused ASM DG (DB) res stop_failed state on secondary node
7026822 acfs_proxy_daemon -the named pipe(/var/cluster/run/oracle_acfs_proxy_fan_pipe_nfsacfsrs) elapsed time
7028417 ACFS housing 10gR2 DB_home in ZC fails to come online after physical node reboot fault
 
        This revision also accumulates feature point patch 145652-09.
 
(from 145335-03)
 
6900551 SUNW.oracle_listener's oracle_listener_stop stop another listener process
6989537 HA-Oracle Dataservice should provide Oracle ASM version information through discovery to DS wizards
6997109 /usr/xpg4/bin/awk within HA-Oracle Dataservice scripts break discovery if invoked by DS wizards
6997111 HA-Oracle Dataservice should provide db_unique_name through discovery to DS wizards
7000009 HA-Oracle Dataservice should provide CRS OCR & Voting location if using ASM to DS wizards
7001510 SUNW.scalable_asm_diskgroup_proxy resource creation fails if multiple ASM diskgroups are specified
7002928 HA-Oracle fails to come online when DB has STATE_DETAILS=Open,Readonly
7003423 HA-Oracle (SUNW.oracle_server) needs to support the Data Guard role SNAPSHOT STANDBY
7005121 HA-Oracle incorrect message format when a Data Guard mis-match role change is detected
7005134 HA-Oracle ASM discovery/validation needs to consider secure usage of /tmp
7007588 clsetup might fail for HA-Oracle configuration creation if LANG=ja
7010458 HA-Oracle/ASM needs to offline the ASM instance or disk group following clrg offline
7013774 oracle_listener_manage method unable to kill listener process
7014053 clsetup ASM wizard doesn't work for 10gR2 (prior 11.2.0.2)
 
(from 145335-02)
 
6947273 resource offline restart dependency from application upon HAStoragePlus RS must be set by clsetup
6978722 ASM validation should be more efficient at setting a GI hard start dependency
6983131 HA-Oracle/ASM 11gR2 incorrect Grid Infrastructure hard start dependency entries for ora.${dg}.dg
 
(from 145335-01)
 
6963569 remove-node from scalable_asm_instance_proxy RG fails if more than one crs_framework version exists
6965972 HA-Oracle/ASM 11gR2 clustered ASM disk group needs to support 3rd-party volume managers
6967714 11gR2 scalable_asm_instance_proxy config validation should have failed when ORACLE_HOME is incorrect
6969233 can't create RAC proxy resource when user has the same group id for Grid and DB home
6969733 OSC 3.3 scproxy_crs_action has logic errors
6974380 RAC wizard fails and results in RAC DB getting stopped when oratab file does not have entry for DB
6975310 providing validation messages|warnings when adding node to scalable_rac_server_proxy RG
6975405 need to move CRS storage proxy dependency creation to the validate method
 
(from 145652-09)
 
6978251 ODG shuts down instance at start up and proxy resource re-attempts to start it every time
7022990 all agents using ds_validate_file or ds_validate_file_owner need largefile support
7024703 SUNW.scalable_rac_server_proxy fails to create when DB name in oratab doesn't match CRS
 
(from 145652-08)
 
7011619 placeholder for FBC 2010/1882 Oracle ACFS Integration
7016918 HA-Oracle/ASM DG resources needs to consider RT version numbers when validating an ASM DG
7017381 zc_sqfs_boot/zc_sqfs_halt fails if line in zc_halt_boot.conf too long or not terminated with newline
 
(from 145652-07)
 
        Feature point patches combine copies of released fixes with new
        features.  This feature point patch ended up containing only copies
        of previously-incorporated fixes with nothing new added.
 
(from 145652-06)
 
6944721 RAC wizard clsetup may hang on different sub wizards for zone cluster
 
(from 145652-05)
 
6477357 fix for 6463264: default property values in RT man pages are now out of date
7013100 deliver data service man page changes for 3.3u1
 
(from 145652-04)
 
6925690 'directory" misspelled as "diectory" in description of some oracle RTs
 
(from 145652-03)
 
6971939 SUNW.scalable_asm_diskgroup_proxy man page is missing
7011126 fix pkgmk warnings
7011278 need man pages for HA-OBIEE
 
(from 145652-02)
 
6965918 DS wizards need to provide support for new 11gR2 SUNW.scalable_asm_diskgroup_proxy RT
 
(from 145652-01)
 
6993431 HA-Oracle validation fails during the resource creation due to ksh issue
 
(from 149491-03)
 
15845249 deliver man pages changes for 3.3u2 data services
 
(from 149491-02)
 
7172829 cmg_register script fails if Oracle Server/Listener and concurrent manager are defined in same RG
7191637 max int value for scalable_asm_diskgroup_proxy probe timeout needs to be increased
7201880 wizard is unable to configure rac-framework RG with udlm for s10 brand zone
 
(from 149491-01)
 
7050551 RAC wizard for Oban config failing reporting incorrectly that SVM is not installed
7086941 udlm (not supported in 4.0) shows up in RAC wizard and blocks rac_framework RG config
7090390 'Unable to determine Oracle CRS version' error after applying patch 145333-09
7115444 HA-Oracle validation error Usage: /bin/su [-] [ username [ arg ... ] ]
7129409 clrs adding START_DEPENDENCIES to ora.DATA.dg via crsctl fails
7129822 HA-Oracle ORA-00119: invalid specification for system parameter LOCAL_LISTENER and ORA-00132 message
7145591 ASM resource scalable_asm_diskgroup_proxy doesn't report error when its ASM DG is gone

--------------------------------
 
There are three (3) possible procedures for installing patches on Oracle
Solaris Cluster. The proper method to use and any additional instructions
for this patch are specified below in the "Special Install Instructions"
section.
 
Refer to the chapter entitled "Patching Oracle Solaris Cluster Software and
Firmware" in the "Oracle Solaris Cluster System Administration Guide" for a
description of the different install processes and instructions on how to
install Oracle Solaris Cluster 3.3 patches.
 
Refer to the man pages for instructions on using 'patchadd' and 'patchrm'
commands provided with Oracle Solaris, to install and remove patches.
 
Read all the instructions in this README before applying or removing
this patch.

-----------------------------
 
NOTE 1:  Install this patch using one of the following patch install
         procedures found in the "Patching Oracle Software and Firmware"
         chapter of the "Oracle Solaris Cluster System Administration Guide:"
 
         * Non Rebooting Patch
         * Rebooting Patch (Node)
         * Rebooting Patch (Cluster and Firmware)
 
NOTE 2:  With the fix for CR 6975405 (need to move CRS storage proxy dependency
         creation to the validate method), the procedure for creating a
         SUNW.scalable_rac_server_proxy resource has changed.
 
         Refer to the Documentation Issues section of the Oracle Solaris
         Cluster 3.3 Release Notes for documentation to support this patch
         with Oracle RAC.
 
NOTE 3:  To get the complete fix for BugId 6965972 (HA-Oracle/ASM 11gR2
         clustered ASM disk group needs to support 3rd-party volume managers),
         follow the procedures located in the Oracle Solaris Cluster 3.3
         Release Notes for "Support for Third-Party Volume Managers with
         Clustered Oracle ASM Disk Groups."
 
NOTE 4:  To get the complete fix for CR6947273 (resource offline restart
         dependency from application upon HAStoragePlus RS must be set by
         clsetup), please also install the following patch:
 
         145333-03 (or greater)  core patch
 
         additionally, please run the following command:
 
         /usr/cluster/lib/ds/serviceregister -f
         /opt/SUNWscor/dsconfig/etc/HAOracleService.xml
 
         After the core patch and this patch are installed, cacao and web
         server should be restarted.
 
NOTE 5:  Revision -06 delivers the same fixes as in update 1 but is not
         equivalent to update 1. Updates can deliver new features that
         will not be in a patch.
 
NOTE 6:  With the fix for CR 7029829 (OSC agent for Oracle DB needs to
         avert race conditions and clear text password), it is possible to
         use Oracle Wallets instead of specifying a user/password within
         the connect_string.
 
         For pre-10gR2, if you do not want to expose the fault monitor
         user and password via the connect_string then Solaris authorization
         should be used and subsequently a slash (/) should be used for the
         connect_string instead of the user name and password. Refer to
         Documentation for Oracle Solaris Cluster Data Service for Oracle Guide.
 
         For 10gR2 or later, you will need to create an Oracle Wallet and
         define credentials as follows on each cluster node where the Oracle
         DB will run:
 
         -bash-3.00$ export ORACLE_HOME=/u01/app/ora11g/product/11.1.0/db_1
         -bash-3.00$ export ORACLE_SID=DB1
         -bash-3.00$ $ORACLE_HOME/bin/mkstore -wrl ${ORACLE_HOME}/network/admin/wallet -create
          Enter password:
 
          Enter password again:
 
         -bash-3.00$
         -bash-3.00$ $ORACLE_HOME/bin/mkstore -wrl ${ORACLE_HOME}/network/admin/wallet -createCredential db1 hauser hauser
          Enter wallet password:
 
          Create credential oracle.security.client.connect_string1
         -bash-3.00$
 
         The location and name of the Oracle Wallet can be any non-volatile
         file location the administrator chooses that is accessible by the
         ${ORACLE_HOME} user. Note, that "db1 hauser hauser" represents
         "SERVICE_NAME user password".
 
         The SERVICE_NAME (db1) needs to reflect the service name that
         identifies the Oracle DB within tnsnames.ora.
 
         ${ORACLE_HOME}/network/admin/sqlnet.ora needs to contain the
         following entries:
 
         -bash-3.00$ more sqlnet.ora
          WALLET_LOCATION =
               (SOURCE =
                       (METHOD = FILE)
                       (METHOD_DATA =
                               (DIRECTORY = ${ORACLE_HOME}/network/admin/wallet)
                       )
               )
 
          SQLNET.WALLET_OVERRIDE = TRUE
         -bash-3.00$
 
         Once the Oracle Wallet and credentials have been defined the
         connect_string must be defined as follows:
 
         -p connect_string=/@${ORACLE_SID}
 
         For example, if ORACLE_SID=DB1 then the connect string would be
 
         -p connect_string=/@DB1
 
         Finally,if Oracle Wallets are being used, the SUNW.oracle_server
         Resource will need an offline restart dependency on the
         SUNW.oracle_listener resource.
 
NOTE 7:  With the fix for 7044445 (errors on the console when unmanage/manage
         an RG containing a SUNW.scalable_acfs_proxy resource), the procedures
         for creating/deleting a SUNW.scalable_acfs_proxy resource for a zone
         cluster have changed.
 
         The resource_dependencies property of the SUNW.scalable_acfs_proxy
         resource must now include the the SUNW.wait_zc_boot resource for the
         zone cluster. Prior deletion of the SUNW.scalable_acfs_proxy resource,
         dependency on SUNW.wait_zc_boot resource must be removed from the
         resource_dependencies property.
 
         After applying the patch, existing SUNW.scalable_acfs_proxy resources
         configured for a zone cluster must be updated to include the
         SUNW.wait_zc_boot resource for the zone cluster in its
         resource_dependencies properties.
 
NOTE 8:  Revision -16 delivers features introduced during update 2.
 
NOTE 9:  The fix for 23228192 (Solaris Cluster HA-Oracle with Data Guard
         starts DB incorrectly) provides a new Active_data_guard extension
         property for SUNW.oracle_server:
 
         Active_data_guard(Boolean)
 
         This extension property indicates whether a physical standby
         instance is additionally managed by Oracle Active Data Guard.
         If this property is set to FALSE, a physical standby database
         is started in MOUNTED mode.  If this property is set to TRUE,
         a physical standby database is started in READ ONLY mode.
 
         With this change, the Type_version of the oracle-server-resource
         must be 9:
 
         # clrs show -p Type_version oracle-server-resource
 
         === Resources ===
 
         Resource: oracle-server-resource
         Type_version:                                    8
 
         If Type_version is less than 9 and Oracle data guard is used,
         do the following:
 
         # clrt register -f /opt/SUNWscor/oracle_server/etc/SUNW.oracle_server SUNW.oracle_server
         # clrs disable oracle-server-resource
         # clrs set -p Type_version=9 -p Active_data_guard=true oracle-server-resource
         # clrs enable oracle-server-resource
 
         If Type_version is less than 9 and Oracle data guard is NOT used,
         do the following:
 
         # clrt register -f /opt/SUNWscor/oracle_server/etc/SUNW.oracle_server SUNW.oracle_server
         # clrs set -p Type_version=9 oracle-server-resource