After apply Oracle 10.2.0.4 CRS PSU bundle 2 , node 1 reboot suddently
OS : hp itanium 11.31 (HP-UX B.11.31 U ia64 3664852670)
Oracle version : 10.2.0.4
After apply 8705958 CRS PSU 2(clusterware patchset update 2) , about one month later , node 1 reboot suddently.
I observed that two logs :
*$ORACLE_HOME/db10g/log/db1/racg/imon_db1.log:*
2010-03-28 09:27:34.857: [ RACG][21] [12463][21][ora.dbrac.db1.inst]: GIMH: GIM-00104: Health check failed to connect to instance.
GIM-00090: OS-dependent operation:mmap failed with status: 12
GIM-00091: OS failure message: Not enough space
GIM-00092: OS failure occurred at: sskgmsmr_13
*/var/adm/syslog/Syslog.log :*
Mar 28 21:38:41 db1 vmunix: file: table is full
Mar 28 21:49:20 db1 syslog: Oracle clsomon failed with fatal status 13.
Mar 28 21:49:20 db1 syslog: Oracle CRS failure. Rebooting for cluster integrity.
Mar 28 21:49:20 db1 vmunix: User requested reset of the system.
Mar 28 21:48:58 db1 vmunix: file: table is full
Mar 28 21:49:20 db1 above message repeats 635 times
Mar 28 21:49:20 db1 vmunix:
Mar 28 21:49:20 db1 vmunix: Oracle CRS TOC for clusterware integrity...
from metalink : File handles not released after upgrade to 10.2.0.3 CRS Bundle#2 or 10.2.0.4 [ID 739557.1]
The article imply that I can apply Patch 7493592 CRS 10.2.0.4 Bundle Patch #2 to fix it , but 8705958 CRS PSU 2 is superset of 7493592
Does anyone have suggestions~? Please kindly let me know
BRS , Jay
--detail steps : apply 8705958 CRS PSU 2(clusterware patchset update 2)
--unzip opatch to latest version 10.2.0.4.9
cd /opt/oracle/crs10g
mv OPatch OPatch.bak
cp -p /opt/oracle_sw/opatch/p6880880_102000_HPUX-IA64.zip /opt/oracle/crs10g/.
unzip /opt/oracle_sw/opatch/p6880880_102000_HPUX-IA64.zip
chown -R oracle:oinstall OPatch
su - oracle
#Verify that the Oracle Inventory is properly configured
/opt/oracle/crs10g/OPatch/opatch lsinventory -detail -oh /opt/oracle/crs10g/
/opt/oracle/crs10g/OPatch/opatch lsinventory -detail -oh /opt/oracle/db10g/
#Unzip the PSE container file
unzip 8705958.zip
#Apply the patch to the CRS Home and all applicable RDBMS homes as root
/opt/oracle/crs10g/OPatch/ocm/bin/emocmrsp
su - root
cd /opt/oracle_sw/PSU/8705958
/opt/oracle/crs10g/OPatch/opatch auto -och /opt/oracle/crs10g/
: /opt/oracle/crs10g/OPatch/ocm/bin/ocm.rsp
Patch node db1? (y/n/abort/N/N1-N2/help):
y
Applying patch 8705958 on node db1
.
.
.
Edited by: user7322888 on 2010/3/30 下午 7:22
Edited by: user7322888 on 2010/3/30 下午 8:23