{"thread":{"id":"4986","subject":"git-clone failures abound","startedAt":"2006-07-26T04:38:46Z","lastAt":"2006-07-26T05:02:54Z","messageCount":2,"participants":["Jeff Garzik","Junio C Hamano"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"24121","messageId":"44C6F1D6.5020208@garzik.org","threadId":"4986","inReplyTo":null,"subject":"git-clone failures abound","fromName":"Jeff Garzik","fromEmail":"jeff@garzik.org","sentAt":"2006-07-26T04:38:46Z","receivedAt":"2006-07-26T04:38:46Z","isPatch":false,"sender":{"key":"jeff@garzik.org","avatar":null},"body":"I've been seeing several reports lately, from different users on various \nLinux platforms, with the same basic bug report\n\n* \"git:// clone of linus's repo times out after 10 minutes\"\n\n* someone says, \"use rsync:// for the initial clone\"\n\n* \"it works, thanks!\"\n\nPeople seems to note that this behavior only started recently.  I wonder \nif linux-2.6.git crossed some sort of size threshold that's too much for \nkernel.org CPU load to bear?  I wonder if git-clone is attempting to \ndelta-ify, when it really should just be sending the objects in bulk?\n\n\tJeff\n"},{"id":"24122","messageId":"7vwta1argh.fsf@assigned-by-dhcp.cox.net","threadId":"4986","inReplyTo":"44C6F1D6.5020208@garzik.org","subject":"Re: git-clone failures abound","fromName":"Junio C Hamano","fromEmail":"junkio@cox.net","sentAt":"2006-07-26T05:02:54Z","receivedAt":"2006-07-26T05:02:54Z","isPatch":false,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"Jeff Garzik <jeff@garzik.org> writes:\n\n> I've been seeing several reports lately, from different users on\n> various Linux platforms, with the same basic bug report\n>\n> * \"git:// clone of linus's repo times out after 10 minutes\"\n>\n> * someone says, \"use rsync:// for the initial clone\"\n>\n> * \"it works, thanks!\"\n>\n> People seems to note that this behavior only started recently.  I\n> wonder if linux-2.6.git crossed some sort of size threshold that's too\n> much for kernel.org CPU load to bear?  I wonder if git-clone is\n> attempting to delta-ify, when it really should just be sending the\n> objects in bulk?\n\nNo, this was all my fault, and sorry about the confusion.\n\nGIT 1.4.1 contains commit 583b7ea3 which implemented a side-band\ncommunication for the upload-pack protocol to give progress bar\noutput to the client downloaders, and in order to do so it\nchanged the pipe structure of the process.  It used to just fork\ntwo processes piped together that exec rev-list and\npack-objects, and exec cleared alarm().  I mistakenly rewrote\nthat part to have an extra process that oversees these two\nprocesses but the overseer does not exec and got killed by\nalarm().  This bug affects the server side.\n\nGIT 1.4.1 was installed on public kernel.org machines and the\nproblem started happening after that.\n\nA fix was relatively simple, and I've issued GIT 1.4.1.1 with it\nlast night -- the master branch also has the same fix.  So when\nthe kernel.org public machines are updated to 1.4.1.1 it should\nsolve the problem.\n"}]}