Date: Tue, 06 Jan 2009 17:30:46 +0000 From: Vincent Hoffman <vince@unsane.co.uk> To: Daniel Bond <db@danielbond.org> Cc: Christopher Arnold <chris@arnold.se>, freebsd-stable@freebsd.org Subject: Re: freebsd-update painfully slow - slower than source code build of world and kernel Message-ID: <49639546.5070608@unsane.co.uk> In-Reply-To: <C68E4D01-D846-4FE4-B71E-9D7ED2EA55E2@danielbond.org> References: <DA7E7739-0631-4B00-8CA8-D8C9E22B2126@danielbond.org> <alpine.BSF.2.00.0901061121530.1996@localhost> <F6385D88-BFAF-47A0-B598-78C971FCBD7C@danielbond.org> <alpine.BSF.2.00.0901061242120.1996@localhost> <08B216B4-79AB-45AB-AB4D-C8CD62196B87@danielbond.org> <alpine.BSF.2.00.0901061731550.1343@localhost> <C68E4D01-D846-4FE4-B71E-9D7ED2EA55E2@danielbond.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Daniel Bond wrote: > The same could be said about CVSup, one could write a caching cvsup > proxy-server, and then we could just get rid of all the other > cvsup-servers, except two (like freebsd-update soon will have). The > point is, for portsnap and freebsd-update to scale properly, it needs > to be opened up to the public, like CVSup is. People running a single > server at home, or maybee two, most like won't want to set up a PROXY > server, and they would be required to update both servers at the same > day for the Proxy server to actually cache something - which many may > not want. And there are a lot of people running a few servers, here > and there. > > > > Sure, a national squid-proxy could work - although, there is no > individual proxy setting for portsnap/freebsd-update.. It honors > HTTP_PROXY environment variable, which a lot of other tools also use. > Some tools might not work via this proxy, especially for local > addresses - the administrators of these servers probably don't want > all the ports tarballs to go via these, and people could use them for > nasty things. So, then we are back to manually setting/specifying the > proxy-server, each time one wants to run the commands - which people > might forget. (Is this getting complicated enough yet..?) We would > basically be creating a whole lot of new potential problems for the > users, to solve the problem in question.. > > > I am also interested in learning how the portsnap protocol works, > maybe there are potential issues with it, that a second eye might > spot, or room for improvement? From what I gather, Colin is a very > cleaver guy, so it is not very likely, but still, other people could > learn from it. > well portsnap/freebsd-update are shell scripts so not too hard to read. The actual transfer protocol is piplined http and is done by /usr/libexec/phttpget (in base so src code available /usr/src/usr.sbin/portsnap/phttpget/phttpget.c ) also see http://www.daemonology.net/phttpget/ > I would like to see these tools as the default recommended tools to > use in the future, and that is why I am so worried about this. > The point I am trying to make is, or actually the question is: Why is > freebsd-update (and portsnap) so secretive? Why can't the average Joe > run his own portsnap-mirror at home? What are we afraid of? I seem to remember once reading that Colin wanted to make it a more polished system before he release it, but i cant find that email anymore. Vince > > I don't see any problems with this, except maybe loosing some detail > in Colin's nice graphs (which would be the case for proxies too). > > > Cheers, > > > Daniel. > > > On Jan 6, 2009, at 5:42 PM, Christopher Arnold wrote: > >> >> >> On Tue, 6 Jan 2009, Daniel Bond wrote: >> >>> reading your answer, you are obviously confusing what I am saying >>> about freebsd-update with the portsnap program. Also, I also wrote >>> in my first post >> No i'm not confusing them, just trying to follow two subjects at the >> same time. Sorry if that is confusing. >> >>> that HTTP_PROXY / Caching proxy server does not help me much. This >>> is because I download a lot of "initial tarball snapshots".. I would >>> rarely see "Cache hits" in my proxy log. I guess I could set >>> something up to fetch nightly via proxy, to keep the data in house, >>> for when I need it. I don't want to use a PROXY server, I feel this >>> is attacking the problem at the wrong end. >>> >> Ok, lets go again. Either you mirror (maybe by having a squid proxy >> and walk the tree) and thats going to me even worse for you. Or you >> use a squid proxy to keep stuff you need close to you and share among >> different installations. >> >> Or you setup one or more national squid proxies and configure your >> machines manually just like you do with cvsup. >> >> >> >>> I agree, I am interested to hear the views of the wise ones. >>> Personally I'm going back to CVSup until freebsd-update and portsnap >>> mirrors are in a more distributed or usable state. >>> >> At least portsnap started to work for me earlier today. Havn't tried >> update yet. >> >> But yes i agree, update and portsnap infrastructure could be done >> better. >> I have some ideas and will try to write them down in a while. >> >> /Chris >> _______________________________________________ >> freebsd-stable@freebsd.org mailing list >> http://lists.freebsd.org/mailman/listinfo/freebsd-stable >> To unsubscribe, send any mail to >> "freebsd-stable-unsubscribe@freebsd.org" > > _______________________________________________ > freebsd-stable@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-stable > To unsubscribe, send any mail to "freebsd-stable-unsubscribe@freebsd.org"
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?49639546.5070608>