{"thread":{"id":"14138","subject":"[BUG] Recent git-svn gets checksum errors on very large fetches","startedAt":"2008-06-25T03:23:57Z","lastAt":"2008-06-25T03:23:57Z","messageCount":1,"participants":["Avery Pennarun"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"81046","messageId":"32541b130806242023r167935f4off85cdd6a49d2750@mail.gmail.com","threadId":"14138","inReplyTo":null,"subject":"[BUG] Recent git-svn gets checksum errors on very large fetches","fromName":"Avery Pennarun","fromEmail":"apenwarr@gmail.com","sentAt":"2008-06-25T03:23:57Z","receivedAt":"2008-06-25T03:23:57Z","isPatch":false,"sender":{"key":"apenwarr@gmail.com","avatar":"https://avatars.githubusercontent.com/u/20592?v=4"},"body":"My svn repo has 17000+ files in 1000+ directories.  I managed to\nnarrow my problem down to this sequence of steps:\n\n    mkdir poo\n    cd poo\n    git init\n    git svn init file:///home/averyp/svn.bak/branches/MyBranch\n    git svn fetch -r3095 2>&1 | tee git.out\n\nWhich culminates in:\n\n    Checksum mismatch: ThirdParty/whateverdir/blahblah\n    expected: c342eaa17fe219a764c06642bb951474\n        got: 848bbb56e4c22755ffb687fb70ae1ac6\n\nFor extra excitement, if I run the same command multiple times, the\nexact file (blahblah) with the checksum mismatch differs each time.\nHowever, it tends to be a file approximately the same \"distance\" down\nin the fetch: it's almost always in whateverdir, which contains only\n333 out of the 17000 files.\n\nThis leads me to believe there's a race condition or buffer problem\nsomewhere causing the problem.  Using git-bisect, I discovered that it\nworks fine in git 1.5.5.1, and that this patch is the one to blame:\n\n  commit ffe256f9bac8a40ff751a9341a5869d98f72c285\n  Author: Adam Roben <aroben@apple.com>\n  Date:   Fri May 23 16:19:41 2008 +0200\n\n    git-svn: Speed up fetch\n\nFigures. :)\n\nReverting it directly seems to cause a conflict, so I can't test for\nsure that the latest master with this patch reverted would work\ncorrectly.\n\nNot quite sure what could be at fault here, or what to do next.\nUnfortunately it's hard to assemble a test case for timing-related\nproblems like this.  Thoughts?\n\nThanks,\n\nAvery\n"}]}