{"thread":{"id":"30482","subject":"Aw: Re: Large repo and pack.packsizelimit","startedAt":"2012-05-09T09:29:05Z","lastAt":"2012-05-09T09:29:05Z","messageCount":1,"participants":["th.acker66@arcor.de"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"191188","messageId":"952610093.980980.1336555745003.JavaMail.ngmail@webmail11.arcor-online.net","threadId":"30482","inReplyTo":null,"subject":"Aw: Re: Large repo and pack.packsizelimit","fromName":"","fromEmail":"th.acker66@arcor.de","sentAt":"2012-05-09T09:29:05Z","receivedAt":"2012-05-09T09:29:05Z","isPatch":false,"sender":{"key":"th.acker@arcor.de","avatar":"https://avatars.githubusercontent.com/u/1358536?v=4"},"body":" \n> On Tue, 8 May 2012, Jeff King wrote:\n> \n> > On Tue, May 08, 2012 at 05:13:13PM -0400, Nicolas Pitre wrote:\n> > \n> > > > This should be fixed in git. Unfortunately, I don't know that it is\n> as\n> > > > trivial as just splitting the incoming stream; we would also have to\n> > > > make sure that there were no cross-pack deltas in the result.\n> > > \n> > > IMHO this is the wrong fix.  The pack size limit was created to deal \n> > > with storage media with limited capacity.  In this case, the repack \n> > > process should be told to limit its memory usage, and pack-index should\n> \n> > > simply be taught to cope.\n> > \n> > Hmm, you're right. I was thinking it helped to deal with memory\n> > addressing issues for 32-bit systems, but I guess\n> > core.packedGitWindowSize should be handling that. IOW, the 10G packfile\n> > should work just fine for normal access.\n> > \n> > However, the OP did say he got an \"out of memory\" error during the\n> > clone. So maybe there is a problem to be fixed in index-pack there.\n> \n> Was the OOM on the remote side (pack-objects) or on the local side \n> (index-pack) ?\n> \n\nTo be exact I did the clone locally on the same machine and so the clone itself worked\nbut I got the OOM during the first fetch. I \"fixed\" this by setting transfer.unpacklimit=100000\nwhich caused only loose objects to be transfered. \nSo in this case I think the OOM was on the remote side. But there is another OOM if I try to repack locally.\nIt seems to me that neither pack-objects nor index-pack respekt pack.packsizelimit and always\ntry to pack all objects to be transferred resp. all local loose objects in one pack.\nI could live wth the transfer.unpacklimit=100000 but the local OOM stops me from using the cloned repo.\n\n\n---\nThomas\n"}]}