Re: 11.1.0.7 patchset experiences

From: LineVoltageHalogen <tropicalfruitdrops_at_yahoo.com>
Date: Tue, 9 Dec 2008 10:07:36 -0800 (PST)
Message-ID: <cd91f6ef-6deb-4a27-8611-cfc4a6e645d6@f3g2000yqf.googlegroups.com>


On Nov 19, 6:59 pm, David <ivoryke..._at_gmail.com> wrote:
> We're running RAC cluster (Dell PowerEdge 1950s w/1 quad core/node) with
> Oracle11 on Linux Intel 5.2 64-bit and are considering the 11.1.0.7
> upgrade. In the next few days the goal is to begin evaluating the
> upgrade in a test environment, but I'd be interested in finding out what
> other people's experience has been.
>
> Any feedback based upon real world experiences with a similar
> configuration regarding the upgrade process itself (including problems
> encountered and the elapsed time to complete the upgrade) and stability
> post upgrade would be most appreciated.
>
> David

We are running Windows 2003 R2 x64 on two Dell 1970s and I have had a few issues.
My main motivation for patching had to do with Oracle chewing up 20-30% cpu when there where
db users connected, it is a known bug. Post upgrade my DBControl was all messed up and I have a new cpu
issue where a java.exe process is taking up to 25% cpu with no users connected. I am currently working with
Oracle support on that issue. Another issue I am having is with our production environment which is a new install from scratch. I tried patching after each component was installed: e.g.) Install Clusterware (11.1.0.6) then patch to 11.1.0.7. The 11.1.0.6 install went off without a hitch, all services were started and no errors were reported. Once I patched the clusterware to 11.1.0.7 I had issues with the *ons* service on one node. There were port issues post upgrade. I worked with Oracle and they gave me a work around (mainly changing the default port number on the second node). So now I go to lay down the ASM binaries (in their own home) and I am prompted with errors on the remote node operation as well as my original port issue with ClusterWare returning (ONS service crashes on the second node and will not start. I am now going to repave my production system and then install 11.1.0.6 (clusterware, asm, rdbms) then go back and patch each component in turn with the hopes that all goes well.

There are not many folks using Oracle on Windows so Oracle support was not aware of the issues I have raised via the three active SR's I have.

Are there any other folks who are running Oracle on Windows who have had problems patching the RAC cluster to 11.1.0.7?

tfd Received on Tue Dec 09 2008 - 12:07:36 CST

Original text of this message