tomcat-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Troy Poppe <t...@usite.net>
Subject Re: request for review: server/config discussion
Date Wed, 21 Jul 1999 03:33:46 GMT
On Tue, Jul 20, 1999 at 08:19:37PM -0700, James Todd wrote:
> > The main issue with a basic config file strategy that I can see is that it
> > should be a requirement to have multiple engines in a cluster running of the
> > same configuration file. Weblogic does this with basic configuration files
> > (it has one server which provides config information to the rest I think).
> > 
> > Tom
> > 

[--snip--]
> regarding clusters reading off of the same config file, and i
> intend to flesh this out a bit more in the "config service"
> section of the discussion doc that i've bounced about, i believe
> a "config service" entity (object, service, whatever) can manage
> propogating config data as sourced from a local file or one
> which was sourced via a url. the only real difference, from
> the reading and processing of the data, is the scheme (local
> or remote file) and how best to determine changes (file mod
> data vs ?someDateMimeHeader?). it may not even be necessary for
> the "config service" to manage a singular file but could
> aggregate such data.
> 
> the external config data/objects as provided by the "config
> service" most likely really shouldn't be exposed to the io
> or xml parsing details.

We also need to consider data that should act as macros for the
config file (ie. server's IP address).

- Troy

Mime
View raw message