{"thread":{"id":"43182","subject":"Re: StGIT repository not clonable?","startedAt":"2006-11-11T03:56:47Z","lastAt":"2006-11-12T15:36:40Z","messageCount":6,"participants":["Karl Hasselström","Petr Baudis","Horst H. von Brand","Catalin Marinas","Jakub Narebski"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"296062","messageId":"200611110359.kAB3ul02013227@laptop13.inf.utfsm.cl","threadId":"43182","inReplyTo":null,"subject":"StGIT repository not clonable?","fromName":"Horst H. von Brand","fromEmail":"vonbrand@inf.utfsm.cl","sentAt":"2006-11-11T03:56:47Z","receivedAt":"2006-11-11T03:56:47Z","isPatch":false,"sender":{"key":"vonbrand@inf.utfsm.cl","avatar":"https://avatars.githubusercontent.com/u/211384?v=4"},"body":"I'm trying to update my StGIT repo here, and get a crash from\ngit-http-fetch (git 1.4.3.4). Trying to clone it anew gives:\n\n\n[vonbrand@laptop13 tmp]$ git-clone http://homepage.ntlworld.com/cmarinas/stgit.git\nerror: Unable to start request\nerror: Could not interpret heads/master as something to pull\n\nWhat am I doing wrong?\n-- \nDr. Horst H. von Brand User #22616 counter.li.org\nDepartamento de Informatica Fono: +56 32 2654431\nUniversidad Tecnica Federico Santa Maria +56 32 2654239\n"},{"id":"293986","messageId":"20061111123634.GD11224@diana.vm.bytemark.co.uk","threadId":"43182","inReplyTo":"200611110359.kAB3ul02013227@laptop13.inf.utfsm.cl","subject":"Re: StGIT repository not clonable?","fromName":"Karl Hasselström","fromEmail":"kha@treskal.com","sentAt":"2006-11-11T12:36:34Z","receivedAt":"2006-11-11T12:36:34Z","isPatch":false,"sender":{"key":"kha@treskal.com","avatar":"https://gravatar.com/avatar/f0120c734b5279b345075a28521e1ac66acb20c9913ffe9bf6ae97e53f7f3f13?d=mp&s=160"},"body":"On 2006-11-11 00:56:47 -0300, Horst H. von Brand wrote:\n\n> I'm trying to update my StGIT repo here, and get a crash from\n> git-http-fetch (git 1.4.3.4). Trying to clone it anew gives:\n>\n> [vonbrand@laptop13 tmp]$ git-clone http://homepage.ntlworld.com/cmarinas/stgit.git\n> error: Unable to start request\n> error: Could not interpret heads/master as something to pull\n>\n> What am I doing wrong?\n\nIt works for me, with\n\n $ git --version\n git version 1.4.3.3.g8387\n\nBut it's horribly slow. Catalin, have you ever packed that repository?\n\n-- \nKarl Hasselström, kha@treskal.com\n"},{"id":"296828","messageId":"b0943d9e0611111359t994d688w9bc6aae8e9183fd3@mail.gmail.com","threadId":"43182","inReplyTo":"20061111123634.GD11224@diana.vm.bytemark.co.uk","subject":"Re: StGIT repository not clonable?","fromName":"Catalin Marinas","fromEmail":"catalin.marinas@gmail.com","sentAt":"2006-11-11T21:59:32Z","receivedAt":"2006-11-11T21:59:32Z","isPatch":false,"sender":{"key":"catalin.marinas@gmail.com","avatar":null},"body":"On 11/11/06, Karl Hasselström wrote:\n> On 2006-11-11 00:56:47 -0300, Horst H. von Brand wrote:\n> > I'm trying to update my StGIT repo here, and get a crash from\n> > git-http-fetch (git 1.4.3.4). Trying to clone it anew gives:\n>\n> It works for me, with\n>\n> $ git --version\n> git version 1.4.3.3.g8387\n>\n> But it's horribly slow. Catalin, have you ever packed that repository?\n\nI've never packed it. IIRC, there was some advise in some GIT document\nor e-mail saying that you shouldn't pack if the export is over a dumb\nprotocol. That's good for people pulling regularly but bad for\ncloning.\n\nAnyway, thanks to Pasky, you can now pull/clone it over the git\nprotocol directly - git://repo.or.cz/stgit.git. This repository is up\nto date. I have a plan to move the main StGIT repository to Pasky's\nserver but I'm a bit busy with other things at the moment.\n\n-- \n"},{"id":"298556","messageId":"20061111223428.GB20192@diana.vm.bytemark.co.uk","threadId":"43182","inReplyTo":"b0943d9e0611111359t994d688w9bc6aae8e9183fd3@mail.gmail.com","subject":"Re: StGIT repository not clonable?","fromName":"Karl Hasselström","fromEmail":"kha@treskal.com","sentAt":"2006-11-11T22:34:28Z","receivedAt":"2006-11-11T22:34:28Z","isPatch":false,"sender":{"key":"kha@treskal.com","avatar":"https://gravatar.com/avatar/f0120c734b5279b345075a28521e1ac66acb20c9913ffe9bf6ae97e53f7f3f13?d=mp&s=160"},"body":"On 2006-11-11 21:59:32 +0000, Catalin Marinas wrote:\n\n> I've never packed it. IIRC, there was some advise in some GIT\n> document or e-mail saying that you shouldn't pack if the export is\n> over a dumb protocol. That's good for people pulling regularly but\n> bad for cloning.\n\nIt's _extremely_ bad when cloning. There's a separate HTTP request for\neach object in the repository, which means lots of time and lots of\nweb server load. It's just about doable to clone the StGIT repository\nnow because it's so small.\n\n-- \nKarl Hasselström, kha@treskal.com\n"},{"id":"297159","messageId":"ej5jt1$9tf$1@sea.gmane.org","threadId":"43182","inReplyTo":"b0943d9e0611111359t994d688w9bc6aae8e9183fd3@mail.gmail.com","subject":"Re: StGIT repository not clonable?","fromName":"Jakub Narebski","fromEmail":"jnareb@gmail.com","sentAt":"2006-11-11T22:48:04Z","receivedAt":"2006-11-11T22:48:04Z","isPatch":false,"sender":{"key":"jnareb@gmail.com","avatar":"https://avatars.githubusercontent.com/u/2706?v=4"},"body":"Catalin Marinas wrote:\n\n> IIRC, there was some advise in some GIT document\n> or e-mail saying that you shouldn't pack if the export is over a dumb\n> protocol. That's good for people pulling regularly but bad for\n> cloning.\n\nBy the way, does dumb protocols download _whole_ packs only? Or do they\ndownload parts of packs (curl can do that, I think)?\n-- \nJakub Narebski\nWarsaw, Poland\nShadeHawk on #git\n\n"},{"id":"294325","messageId":"20061112153640.GD7201@pasky.or.cz","threadId":"43182","inReplyTo":"ej5jt1$9tf$1@sea.gmane.org","subject":"Re: StGIT repository not clonable?","fromName":"Petr Baudis","fromEmail":"pasky@suse.cz","sentAt":"2006-11-12T15:36:40Z","receivedAt":"2006-11-12T15:36:40Z","isPatch":false,"sender":{"key":"pasky@ucw.cz","avatar":"https://avatars.githubusercontent.com/u/18439?v=4"},"body":"On Sat, Nov 11, 2006 at 11:48:04PM CET, Jakub Narebski wrote:\n> Catalin Marinas wrote:\n> \n> > IIRC, there was some advise in some GIT document\n> > or e-mail saying that you shouldn't pack if the export is over a dumb\n> > protocol. That's good for people pulling regularly but bad for\n> > cloning.\n> \n> By the way, does dumb protocols download _whole_ packs only? Or do they\n> download parts of packs (curl can do that, I think)?\n\ncurl can, but it might very easily get even much more expensive than\ndownloading the whole patch unless your latency is very small and\nbandwidth very tight, which would be quite a unusual situation.\n\nIt's true that repacking can hurt dumb protocols - if you repack often,\ndumb clients will have to re-fetch the single whole patck with all the\nstuff they already have plus the few additional objects they are\nmissing.\n\nBut at least packing once can be a huge improvement and won't hurt the\ndumb clients since their problem is with incremental fetches.\nFurthermore, if you do just repack, not repack -a, the cost for dumb\nprotocols is quite small (though it's not optimal packing strategy):\n\nIt is not unlikely at all that if you have set of unpacked objects A,\nclient fetches that, then you create set of objects B and then repack,\ncreating pack(A \\cup B), this pack will still be much smaller than the\nset of objects B (even if |A| >> |B|) so it's more beneficial even for\nthe dumb clients to refetch the A objects contained in the pack, instead\nof fetching just the unpacked B objects.\n\nBy the way, in case of glibc-cvs the pack sice is 104M, and after\nimporting new CVS changes after few days, the repository size doubled to\n200M. git-repack -a -d brought that _back_ to 104M!\n\nPacks are a funny thing.\n\n-- \n\t\t\t\tPetr \"Pasky\" Baudis\nStuff: http://pasky.or.cz/\n#!/bin/perl -sp0777i