What’s new in Oracle 11.2.0.2

Oracle has recently released 11.2.0.2 for Solaris (x86, SPARC), Linux, Aix, HP-UX (IA).

I’ve tried installing it on one of my Solaris boxes. Following are few Solaris specific observations:

-> 11.2.0.2 is a full install & not a patch. One don’t need any existing installation to apply/install 11.2.0.2. Size of this patch is approximately 7G and is currently available on Metalink for download. 

This patch is available in multiple zip files (6 for Solaris x86 and 7 files for SPARC). All the files may not be required for install.

p10098816_112020_platform_<1,2>of6.zip - Oracle Database
p10098816_112020_platform_3of6.zip - Oracle Grid Infrastructure
Rest of the files contain client, deinstall, examples.

 -> Oracle now allows 2 type of upgrades

a.  In-Place Upgrade - Legacy method of replacing the existing binaries
b.  Out-of-Place Upgrade – Install new home and the upgrade the database. This approach needs more space, as we’ll have old and new software present at  same time. This approach need less downtime.

When I tried first approach of “In-place migration” to upgrade grid home, I faced INS-32025 error. Didn’t get change to explore the reason behind it. But it enforces to go for out-of place approach.

-> With release of 11.2.0.2, ACFS is now supported on Solaris/Aix (earlier it was only supported on Linux)

When running root.sh, we’ll see

ACFS-9200: Supported
ACFS-9300: ADVM/ACFS distribution files found.
ACFS-9307: Installing requested ADVM/ACFS software.
updating /platform/i86pc/boot_archive
ACFS-9308: Loading installed ADVM/ACFS drivers.
ACFS-9327: Verifying ADVM/ACFS devices.
ACFS-9309: ADVM/ACFS installation correctness verified.
……………
……………
……………
CRS-2672: Attempting to start ‘ora.registry.acfs’ on ‘solsrv-20′
CRS-2676: Start of ‘ora.registry.acfs’ on ‘solsrv-20′ succeeded

-> New resource ora.registry.acfs (Oracle ACFS registry resource) is added

 
-> New resource ora.cvu (Cluster Verification Utility) is added

This resource invokes cluster verification utility (I guess it is executed after every 6 hours) and store it’s output in $GRID_HOME/log/nodename/cvu/cvulog/cvu20101101111536.log

Detail trace of this utility is also saved under  $GRID_HOME/log/nodename/cvu/cvutrc

-rw-r–r–   1 oragrid  oinstall 5720101 Nov  1 11:16 cvutrace.log.0
-rw-r–r–   1 oragrid  oinstall 26213985 Nov  1 05:15 cvutrace.log.3

—– snip from the cvu20101101111536.log

SYS::OVERALL_STATUS::VFAIL::System
SYS::NODE_STATUS:solsrv-21::WARNING
MSG_START
PRVF-7543 : OS Kernel parameter “tcp_smallest_anon_port” does not have proper value on node “solsrv-21″ [Expected = “9000” ; Found = “32768”].
PRVF-7543 : OS Kernel parameter “tcp_largest_anon_port” does not have proper value on node “solsrv-21″ [Expected = “65500” ; Found = “65535”].
PRVF-7543 : OS Kernel parameter “udp_smallest_anon_port” does not have proper value on node “solsrv-21″ [Expected = “9000” ; Found = “32768”].
PRVF-7543 : OS Kernel parameter “udp_largest_anon_port” does not have proper value on node “solsrv-21″ [Expected = “65500” ; Found = “65535”].
MSG_END
SYS::NODE_STATUS:solsrv-20::VFAIL
MSG_START
PRVF-7573 : Sufficient swap size is not available on node “solsrv-20″ [Required = 16GB (1.6777216E7KB) ; Found = 15.6272GB (1.6386296E7KB)]
PRVF-7543 : OS Kernel parameter “tcp_smallest_anon_port” does not have proper value on node “solsrv-20″ [Expected = “9000” ; Found = “32768”].
PRVF-7543 : OS Kernel parameter “tcp_largest_anon_port” does not have proper value on node “solsrv-20″ [Expected = “65500” ; Found = “65535”].
PRVF-7543 : OS Kernel parameter “udp_smallest_anon_port” does not have proper value on node “solsrv-20″ [Expected = “9000” ; Found = “32768”].
PRVF-7543 : OS Kernel parameter “udp_largest_anon_port” does not have proper value on node “solsrv-20″ [Expected = “65500” ; Found = “65535”].
MSG_END
CTSS::OVERALL_STATUS::SUCC::Oracle Cluster Time Synch Service
CTSS::NODE_STATUS:solsrv-21::SUCC
CTSS::NODE_STATUS:solsrv-20::SUCC
OHASD::OVERALL_STATUS::SUCC::Oracle High Availability Daemon
OHASD::NODE_STATUS:solsrv-21::SUCC
OHASD::NODE_STATUS:solsrv-20::SUCC
CRS::OVERALL_STATUS::SUCC::Oracle Clusterware
CRS::NODE_STATUS:solsrv-21::SUCC
CRS::NODE_STATUS:solsrv-20::SUCC
SCAN::OVERALL_STATUS::SUCC::Scan
SCAN::NODE_STATUS:localnode::SUCC
SCAN::NODE_STATUS:solsrv-21::SUCC
SCAN::NODE_STATUS:solsrv-20::SUCC
NODE_APP::OVERALL_STATUS::SUCC::Node Apps
NODE_APP::NODE_STATUS:solsrv-21::SUCC
NODE_APP::NODE_STATUS:solsrv-20::SUCC

 

-> Few new directories in log

drwxr-x—   2 root     oinstall     512 Oct 21 21:22 crflogd
drwxr-x—   2 root     oinstall     512 Oct 21 21:22 crfmond

And also a new process

# ps -ef | grep crf
root   845     1   0   Oct 25 ?          75:11 /oragrid/11.2/bin/ologgerd -M -d /oragrid/11.2/crf/db/solsrv-21

Seems like Cluster Health Monitor.

-> Installer now check the consistency of the /etc/resolve.conf on all cluster nodes.

If it finds mismatch, it will fail the check. Secondly, resolve.conf cannot have both “search and domain” entries.

-> Installer check for specific ASM disk ownership

It expects ownership as oragrid:asmadmin (grid owner and asmadmin group)

-> Few more tcp/udp related parameters are required

tcp_smallest_anon_port
tcp_largest_anon_port
udp_smallest_anon_port
udp_largest_anon_port

-> Installer also expects  LoginGraceTime set to 0

changes required in /etc/ssh/sshd_config

About these ads
This entry was posted in Oracle 11.2.0.x and tagged , . Bookmark the permalink.

One Response to What’s new in Oracle 11.2.0.2

  1. Pingback: 2010 in review | Oracle-Hands-On

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s