<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
<title></title>
</head>
<body bgcolor="#ffffff" text="#000000">
Chuck,<br>
<br>
We are attempting to do just that here at LBNL. The "APPROVAL"
setup in RT does not seem to create the kind of audit trail we would
like. We would like a request to keep it's number, that way every
stage, from approval for work to approval for Acceptance testing to
approval for installation is in history for the same request numer and
therefore easier to track. If RT would allow for a few more options in
the drop-down selection for staus (like "pending approvel", "approved
for work", "approved for QA", "approved for inplementation"), we could
write scrips to move a request from a particular approval queue to a
work queue and back to another approval queue (for QA as an example)
and on to implemntation. That way queues can be used as steps in a work
flow and the history of a request remains intact with the same number,
going from queue to queue in the workflow. Well, anyway, that's one
idea we have. It may require scrips that modify the status field
temporarily until the request can be moved to another queue and modify
the status again. We don't know yet because we are currently in the
process of playing with this design idea, but at least it is an idea
that could work, provided RT can handle the change to the status field.<br>
<br>
Kenn<br>
<br>
Chuck Boeheim wrote:<br>
<blockquote cite="mid4416F775.8020307@slac.stanford.edu" type="cite">Hi
Folks,
<br>
Has anyone come up with any good methodologies for change control
<br>
of an RT installation? Many of the components are database records
<br>
that don't seem amenable to CVS control, scrips and templates as well
<br>
as custom field definitions and queue setups. We'd like to have a
record
<br>
of what changed when and by whom, and be able to reconstruct the
<br>
state of the system or back out a change. Even an audit record of
what
<br>
changed would be better than nothing, even if it can't be used directly
<br>
for backing out a change. Has anyone tackled this?
<br>
Best,
<br>
-Chuck Boeheim
<br>
Stanford Linear Accelerator Center
<br>
<pre wrap="">
<hr size="4" width="90%">
_______________________________________________
<a class="moz-txt-link-freetext" href="http://lists.bestpractical.com/cgi-bin/mailman/listinfo/rt-users">http://lists.bestpractical.com/cgi-bin/mailman/listinfo/rt-users</a>
Community help: <a class="moz-txt-link-freetext" href="http://wiki.bestpractical.com">http://wiki.bestpractical.com</a>
Commercial support: <a class="moz-txt-link-abbreviated" href="mailto:sales@bestpractical.com">sales@bestpractical.com</a>
Discover RT's hidden secrets with RT Essentials from O'Reilly Media.
Buy a copy at <a class="moz-txt-link-freetext" href="http://rtbook.bestpractical.com">http://rtbook.bestpractical.com</a>
We're hiring! Come hack Perl for Best Practical: <a class="moz-txt-link-freetext" href="http://bestpractical.com/about/jobs.html">http://bestpractical.com/about/jobs.html</a></pre>
</blockquote>
</body>
</html>