www-mirrors mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From jason andrade <ja...@dstc.edu.au>
Subject Re: cvs commit: site/xdocs/dev mirrors.xml
Date Wed, 27 Nov 2002 23:38:59 GMT
On Wed, 27 Nov 2002, Scott Kveton wrote:

> > 4 might be a bit small, but i guess there might be some policy there.  in the
> > US it might make sense to have 15 mirrors.  in australia 6.  in Tibet, 2.  it
> > would be a (clearly understood) function of network and demand (number of
> > internet users) in that region.
>
> I think 4 is pretty arbitrary ... how do we keep the 4 best mirrors?  I
> don't think you should have a million mirrors; keep it as several
> high-quality mirrors ... of course how do you make sure they are high
> quality?

just setup a table to start with with definitions and build on that as time
passes.  the US starts with 16, etc.

high quality is of course going to be subjective.  10Mbps is not seen as
high quality in the US but it certainly would be in Djibouti.  of course
this leads into also needing to have something similar to the qmail check
or the CPAN check - a file with a timestamp written into the apache top
level directory with a master process that checks to see how up to date
a mirror is.

i am pretty sure there is a CPAN person here (Bjorn?) who might be able
to say whether we could use the CPAN checking system for apache mirrors.

it is still a manual process i understand.  all the check does is let the
person (Jarkko?) know if mirrors are falling out of date and if he gets
3 or 4 messages then he contacts the mirror admin directly about it.


regards,

-jason


Mime
View raw message