Kam Leo wrote:
On 7/20/07, Timothy Murphy <tim@xxxxxxxxxxxxxxxxxxxxxx> wrote:
I asked a while ago if there was any way
of only downloading a package once
and then installing it on several machine,
and I was advised that the yum-downloadonly plugin
would serve this purpose.
I just got round to looking at this,
and I don't really see how to set this up.
What I would like is an option to yum, say localrepo,
so that if I say "yum --localrepo update" then
1. yum will look first to see if the package it wants
is already in the directory /common/RPMS/
(specified in some config file),
and install it from there if it is found; and
You do not need the yum-downloadonly plugin. Edit /etc/yum.conf and
change "keepcache=0" to "keepcache=1". All the rpms that yum downloads
will remain in the /var/cache/yum directories (separate directory for
each repo) until you dispose of them (yum clean packages, rm xxx, mv
yyy, etc.).
I've been keeping the cache, but now with so many updates, I've got a
bunch of older versions which have newer packages also downloaded. Is
there a utility which will cleanup the older packages and leave the
newer ones? If I can find that, I sure I could then devise a way to
combine yum, cron, "that" utility, rsync, createrepo, and the priority
plugin for yum to download the minimal amount of stuff for multiple
servers/desktops/virtual machines (right now I only have one, and this
problem is one of a couple reasons why I haven't expanded yet).
2. If it does not find the package locally
it will download and install it,
and save a copy in /common/RPMS .
Is this already available, eg using yum localinstall ?
--
Timothy Murphy
