Re: arbitration of multiple master_sites

2008-05-18 Thread markd
>>> So if we had this master_sites keyword in port foo: >>> >>> master_sites \ >>> sourceforge \ (ping resp t9) >>> ftp://example1.org \(ping resp t4) >>> http://example2.org \ (ping resp t7) >>> freebsd (ping resp t1) >> >> Actuall

Re: arbitration of multiple master_sites

2008-05-18 Thread Joshua Root
Rainer Müller wrote: > [EMAIL PROTECTED] wrote: -Sourceforge, gnu, etc., special mirrors go after sorted url list >>> No. They get mixed in with the other master_sites and sorted. >> >> So if we had this master_sites keyword in port foo: >> >> master_sites \ >> sourceforge \

Re: arbitration of multiple master_sites

2008-05-17 Thread Rainer Müller
[EMAIL PROTECTED] wrote: >>> -Sourceforge, gnu, etc., special mirrors go after sorted url list >> No. They get mixed in with the other master_sites and sorted. > > So if we had this master_sites keyword in port foo: > > master_sites \ > sourceforge \ (ping resp t9) > ftp

Re: arbitration of multiple master_sites

2008-05-17 Thread markd
>> -Sourceforge, gnu, etc., special mirrors go after sorted url list > >No. They get mixed in with the other master_sites and sorted. So if we had this master_sites keyword in port foo: master_sites \ sourceforge \ (ping resp t9) ftp://example1.org \(ping resp t4)

Re: arbitration of multiple master_sites

2008-05-16 Thread Joshua Root
[EMAIL PROTECTED] wrote: > Question for [EMAIL PROTECTED] Is this an accurate description of the fetch > behavior > now? > > -Ping the list of master_sites and sort them based on ping response times Right. > -Sourceforge, gnu, etc., special mirrors go after sorted url list No. They get mixed

arbitration of multiple master_sites

2008-05-16 Thread markd
>> I always assumed that the master_sites were tried in order. I'd like >> to >> document the true behavior in the guide. So it does a ping test of >> all >> master_sites and uses the one with the fastest ping response? >> > >Its a recent feature by jmr: > >http://trac.macports.org/ticket/1489

Re: arbitration of multiple master_sites

2008-05-16 Thread William Siegrist
On May 16, 2008, at 3:12 PM, [EMAIL PROTECTED] wrote: master_sites lists . has a lower ping for the reporter, hence it was used. I always assumed that the master_sites were tried in order. I'd like to document the true behavior in the guide. So it does a ping test of all master_sit

arbitration of multiple master_sites

2008-05-16 Thread markd
> master_sites lists . >has a lower ping for the reporter, hence it was used. I always assumed that the master_sites were tried in order. I'd like to document the true behavior in the guide. So it does a ping test of all master_sites and uses the one with the fastest ping response? Mark