« Running Autoconfig in Test mode - Useful for Making 11i Rapid Clones faster.. | Main | Quick script for comparing two init.ora files »

Reducing time on txkWfClone.sh while running Autoconfig

Preface

Running Autoconfig on multiple tiers can take a long time sometimes and many a times, the culprit is txkWfClone.sh

The real deal

During cloning, execution of txkWfClone  is required only for the first autoconfig run. For subsequent Autoconfig  runs,  it is not required.

How can you be smart?

There are two smart things you can do to help yourself and make this happen.

1). Move the phase of the script txkWfClone.sh to INSTALL_SETUP phase in $FND_TOP/admin/driver/fndtmpl.drv:

Open the file $FND_TOP/admin/driver/fndtmpl.drv and make the following changes:

Before:
fnd admin/template txkWfClone.sh    INSTE8_SETUP     
<s_com>/admin/install/<s_contextname> txkWfClone.sh 700

After:
fnd admin/template txkWfClone.sh    INSTALL_SETUP     
<s_com>/admin/install/<s_contextname> txkWfClone.sh 700


2). Replace %s_webhost% with %s_webentryhost% and %s_domainname% with %s_webentrydomain%, in the template $FND_TOP/admin/template/txkWfClone.sql

How does the savings add up..

Changing the autoconfig phase of this script from  INST8_SETUP to INSTALL_SETUP will prevent it from unwanted execution, resulting in better performance for future autoconfig runs.

The role of txkWfCLone.sh is to create workflow related references in  db tier of target system in the cloning process.

During txkWfClone.sh's execution, the comparison between source system and target system URLs is done.  Essentially, in a scenario where we are creating an additional middle tier from an existing  middle tier/(s) having one common database tier, it will result in  redundant operation of changing/creating workflow references for target  middle tier.

For comparison, txkWfClone constructs an url for target system using  the context variables as follows:

'%s_webentryurlprotocol%'||'://'||'%s_webhost%'||'.'||'%s_domainname%'||':'||'
%s_active_webport%'||'/pls/'||'%s_dbSid%'

Because the url for target system is constructed using s_webhost  context variable, the script creates distincts urls for each middle tier  which is not necessary because they use the same database instance and hence the workflow references are the same.

This results into a good amount  of time consumption for every  autoconfig run while cloning.

This can be avoided by changing the use of %s_webhost% with the use of %s_webentryhost%  in the template txkWfClone.sql.

This way the urls created for each middle  tier would be the same as it should be in the case of a system with multi apps tiers and one db instance.

Conclusion

It is good to save time during cloning and txkWfClone.sh can be easily fixed to avoid duplicate effort. Simple fixes like this should be absorbed into regular cloning practices for saving time and effort.

TrackBack

TrackBack URL for this entry:
http://blogs.oracle.com/mte1521/mt-tb.cgi/2120

Comments (2)

Good one, always enjoy reading your blogs.
I was told the latest adconfig patch will be changed to include to run only during setup phase.

Cheers.

Michael:

I used the options mentioned and saved around 5 mts per node. We refresh a 5 node RAC environment, and resulted in 25 mts of refresh time. From 3hrs to 2hrs 35mts. Thanks for the information.

Post a comment

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)

About This Entry

This page contains a single entry from the blog posted on January 8, 2008 1:07 PM.

The previous post in this blog was Running Autoconfig in Test mode - Useful for Making 11i Rapid Clones faster...

The next post in this blog is Quick script for comparing two init.ora files.

Many more can be found on the main index page or by looking through the archives.

Top Tags

Creative Commons License
This weblog is licensed under a Creative Commons License.
Powered by
Movable Type and Oracle