>  I'm not sure why you would want yet another way to do it.

Basically for migration and maintenance purposes

1) I've got a set of scripts that can populate my Hudson jobs. So I can
delete all the jobs, re-run my scripts and nearly everything is back to
where it was.

I took those scripts, re-did the underlying xml manipulation classes to
work against a Jenkins server, and now I can re-constitute my jobs onto
Jenkins.

2) As a nice side-effect, I wrote the scripts to warn me when the expected
configuration is different from the actual configuration. I can find out if
something got changed "by accident" and if I want I can either revert it on
the Server or update my scripts to make it permanent

3) Started off with 15 jobs and because I scripted the heck out of the
thing, I was able to create lot's of little jobs at will (I currently have
750 jobs).

So: adding a job or jobs is trivial, doing bulk changes is trivial, etc.
The scripts opened up a lot of possibilities for doing things in
Hudson/Jenkins that would have been overwhelming to do manually.

The effort to write the scripts was initially high , but over time, the
maintenance effort has become puny compared to what it would have been.

4) the scripts are in source control. So I can revert to a previous Jenkins
configuration, by:

- reverting the scripts to a previous revision

- re-running them

This process only takes around 10-15 minutes compared to ?? minutes to do
it manually.


John

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/groups/opt_out.


Reply via email to