[imp] What's the advantage of using CVS?

[Eric Rostetter]

Quoting Craig White <craigwhite at azapple.com>:

> Not a stupid question.

Agreed.
 
> Production environments should almost always stay with stable releases.


In general true, but many find the need to run CVS RELENG releases
in production, and if done carefully, this is generally a good thing.
But this does take extra work/effort, and isn't for the lazy or inexperienced.

> Using CVS releases enable you to participate in the software development
> phase by working through the changes & problems and reporting back.

Or, just to benefit from bug fixes (and new features or enhancments) made
between releases.
 
> It's been my experience that the past 6 months, it's been acceptable to
> use Horde RELENG_2 / IMP RELENG_3 in a production environment but YMMV

Yes, generally so, if you are diligent.  Don't do this though until you 
are comfortable with CVS and how it works, and have signed up for the
cvs at lists.horde.org mailing list.

> But doing this means that I monitor the mail lists and don't update the
> cvs unless I've got time to monitor, test & fix.

Exactly.  And have a good understanding of how CVS works (updates/merges,
conflicts, etc).  Knowing how to revert changes would also be a good
thing to know before doing this.

You also need to understand that CVS won't change your configuration
files, and that you may need to manually merge configuration changes
in.

I really like running CVS RELENG code.  But you do need to know some
stuff to do so successfully.  It isn't as hard as it sounds, but you
might want to play with it first for a while, then move to using it
in production once you feel confortable with it.

But, it is never recommended to run CVS HEAD in production (yes, many
do, including me, but it is not recommended, and takes some serious 
effort, and causes serious problems).

> also...
> 
> All you need to know about using Horde/IMP etc CVS is here...
> <http://www.horde.org/source/>

And in the faq at http://www.horde.org/faq/admin/

-- 
Eric Rostetter
The Department of Physics
The University of Texas at Austin

Why get even? Get odd!