archiva-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jeff Jensen" <jeffjen...@upstairstechnology.com>
Subject Clarifications on DB, container, and M1 usage
Date Fri, 09 Nov 2007 04:49:40 GMT
Hi,

I am about to setup the beta 3 to determine feasibility of switching from
Proximity (I did a quick setup exploration so far).  I have been reviewing
archive messages on user and dev, and read the wiki for education on
Archiva.  I've also reviewed many of the JIRAs to understand current known
issues.

I read Joakim's plans on docs on the dev list.  Looks good - I think they
will answer many of my questions when complete... :-)

I am wondering about advice for these things if anyone has a moment to
share:

 - What are decision criteria to use Derby or use MySQL/Postgres, etc.
instead?  I know part of the full answer is not Archiva related, and in
asking I am wondering specifically about Archiva use cases that would
warrant "not Derby"?  That includes upgrade pros and cons.

 - Plexus container vs Tomcat: Having never used Plexus before...  getting
going with it was very easy - couldn't be easier.  I am wondering about
stability/reliability with "production usage".
After some initial attempts, I don't have Tomcat working yet.  What makes it
worth the extra effort for Tomcat setup/when is it advisable to not use the
Plexus container?
[Another concern is the Tomcat instructions are a very non-standard
configuration (I found the JIRA issue on its config instructions, which I
agree with).  I wonder if a more standard configuration could work or not.
Note also that Tomcat 6 has different dir structure, requiring additional
doc updates over the existing corrections]

 - Maven 1: I didn't see much for notes or archive emails on using Archiva
with M1, e.g. Archiva config and features.  On the Getting Started page's
"Setting up your Archiva instance" section, it says "it can be both an m1
and an m2 repository."  I'm not sure what this means - does this mean I can
define all my repos as M2 format and Archiva will automatically serve M1 and
M2 clients, transparently converting in real time?  If so, that is a
fantastic feature; and would deploys work the same?




Mime
View raw message