Re: Measuring testing vs stable vs oldstable usage from security mirrors logs
On 08/08/13 at 23:21 +0200, Kurt Roeckx wrote:
> > For your info, using ftp.debian.org logs on klecker.debian.org:
> > http://people.debian.org/~spaillard/mirrors/ftp.debian.org.statistics
> >
> > The perl script use to parse Apache logs:
> > http://people.debian.org/~spaillard/mirrors/analyze-debian-mirror.pl
> >
> > It still lacks checking unique IP when counting hits.
>
> That script generates this for ftp.be.debian.org, 1 week log starting the 28th of
> july.
Kurt kindly sent me this log.
487 distinct IPs retrieved a Packages.(gz|bz2) file in /debian/.
wheezy 292 IPs, 60%
jessie 58 12%
sid 52 11%
squeeze 203 42%
The sum is > 100%, some IPs retrieved several Packages files. If we
consider only the higher release an IP retrieves, I get:
sid 52 11%
jessie 41 9%
wheezy 262 54%
squeeze 132 27%
Lucas
Reply to: