Creating a local RPM repository

Timothy Murphy gayleard at eircom.net
Sun Nov 8 00:13:23 UTC 2009


Tim wrote:

> Personally, I find the simple HTTP/FTP caching approach with Squid is
> the simplest:  You configure your yums, on all machines, to use just one
> mirror, and to fetch through your proxy.  Squid caches what you get.
> And you only download, and cache, the packages that you actually use.

You convinced me to start squid,
but unfortunately after reading my trusty tutorial,
<http://www.brennan.id.au/11-Squid_Web_Proxy.html>,
and looking through /etc/squid/squid.conf ,
I decided the chances of my making a mistake,
and cutting off my family from the internet,
was too high to risk.

I do realise that it would be good to run squid on my server,
but as I said it seems a risky enterprise.

Is it possible to use squid just for yum, say,
as an experimental start?

> Some of the local mirroring options involve blindly downloading every
> update that's released.  Whether, or not, you use that package.  For me,
> that'd be a huge waste of bandwidth and drive space.

That was exactly what I felt about setting up a mirror,
which as far as I could see meant mirroring the official repository.


-- 
Timothy Murphy  
e-mail: gayleard /at/ eircom.net
tel: +353-86-2336090, +353-1-2842366
s-mail: School of Mathematics, Trinity College, Dublin 2, Ireland




More information about the fedora-list mailing list