Re: reuse apt-get cache

[Benedict Verheyen <linux4bene@pandora.be>]

Op di 23-09-2003, om 22:17 schreef Osamu Aoki:
> Hi, my neibour :-)
> 
> On Tue, Sep 23, 2003 at 06:02:08PM +0000, Benedict Verheyen wrote:
> > i have a server and it has a quite extensive sources.list. Now i would 
> > like to make a cron job that does "apt-get update" every so often and 
> > then use that cache for my connected client so i don't have to download 
> > the packages list again and thus use unnecessary bandwidth.
> 
> 1) Why do "apt-get update", do "apt-get update; apt-get -d upgrade"
> 2) web cache ==>  use squid
> 3) Always install not by CRON.

Well, the basic idea is to save bandwith and only have 1 computer get
the packages lists. As other people have suggested apt-proxy seems to be
able to do that.

> But why have quite extensive sources.list for server.  You should be
> running STABLE ! for server.

I know. I like living on the edge :) and it's not a production server.
But i know i should run stable.

> 
> > In the future i plan to make packages of my own scripts. Can i combine
> > those? I mean putting my packages in the cache of packages on the server
> > and thus also have them available for the pc's connected to my LAN?
> 
> You can do this in many ways.  But I found running squid to be most
> simple and flexible.

Another suggestion posted here is to make my own repository with
dpkg-scanpackages and adding this to the sources lis of the server.

Thanks all for the suggestions made!
Benedict