{"thread":{"id":"10322","subject":"linux-2.6.git mirror","startedAt":"2007-10-16T20:27:09Z","lastAt":"2007-10-19T22:50:17Z","messageCount":6,"participants":["Medve Emilian-EMMEDVE1","Linus Torvalds"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"56118","messageId":"598D5675D34BE349929AF5EDE9B03E2701684C77@az33exm24.fsl.freescale.net","threadId":"10322","inReplyTo":null,"subject":"linux-2.6.git mirror","fromName":"Medve Emilian-EMMEDVE1","fromEmail":"emilian.medve@freescale.com","sentAt":"2007-10-16T20:27:09Z","receivedAt":"2007-10-16T20:27:09Z","isPatch":false,"sender":{"key":"emilian.medve@freescale.com","avatar":null},"body":"Hi Linus,\n\n\nI'm trying to setup a mirror of your Linux tree with git 1.5.3.1 and\nv2.6.23-5054-g821f3ef and I get the following error during this\nscenario:\n\n$ mkdir linux\n$ cd linux\n$ git --bare init --shared=all\nInitialized empty shared Git repository in /home/emmedve1/linux/\n$ git remote add --mirror -f origin\ngit://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git\n$ git fsck --full --strict\t<- all fine\n$ git remote update\nUpdating origin\nerror: Object 5dc01c595e6c6ec9ccda4f6f69c131c0dd945f8c is a tree, not a\ncommit\nerror: Object 5dc01c595e6c6ec9ccda4f6f69c131c0dd945f8c is a tree, not a\ncommit\n$ git fsck --full --strict\t<- all fine\n\nThe situation is similar with the git tree:\n\nerror: Object a0e7d36193b96f552073558acf5fcc1f10528917 is a blob, not a\ncommit\n\nOther repositories (sparse for example) don't necessarily expose this,\nbut I didn't test too many. This situation doesn't seem to surface for\nnon-bare clones. I searched a bit the web for this, but nothing obvious\nshowed up.\n\nIs this something I should be worried about?\n\n\nThanks for your patience,\nEmil.\n"},{"id":"56122","messageId":"598D5675D34BE349929AF5EDE9B03E2701684C89@az33exm24.fsl.freescale.net","threadId":"10322","inReplyTo":"598D5675D34BE349929AF5EDE9B03E2701684C77@az33exm24.fsl.freescale.net","subject":"RE: linux-2.6.git mirror","fromName":"Medve Emilian-EMMEDVE1","fromEmail":"emilian.medve@freescale.com","sentAt":"2007-10-16T20:47:49Z","receivedAt":"2007-10-16T20:47:49Z","isPatch":false,"sender":{"key":"emilian.medve@freescale.com","avatar":null},"body":"> I'm trying to setup a mirror of your Linux tree with git 1.5.3.1 and\n> v2.6.23-5054-g821f3ef and I get the following error during this\n> scenario:\n\nInstead of v2.6.23-5054-g821f3ef please read v1.5.3.4-206-g58ba4f6.\n"},{"id":"56434","messageId":"alpine.LFD.0.999.0710181518120.26902@woody.linux-foundation.org","threadId":"10322","inReplyTo":"598D5675D34BE349929AF5EDE9B03E2701684C77@az33exm24.fsl.freescale.net","subject":"Re: linux-2.6.git mirror","fromName":"Linus Torvalds","fromEmail":"torvalds@linux-foundation.org","sentAt":"2007-10-18T22:26:08Z","receivedAt":"2007-10-18T22:26:08Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Tue, 16 Oct 2007, Medve Emilian-EMMEDVE1 wrote:\n>\n> $ git remote update\n> Updating origin\n> error: Object 5dc01c595e6c6ec9ccda4f6f69c131c0dd945f8c is a tree, not a commit\n> error: Object 5dc01c595e6c6ec9ccda4f6f69c131c0dd945f8c is a tree, not a commit\n\nInteresting. Something seems to be assuming that all tags are commits. \nWhich is not true. You can have (and the kernel repo does this) a tag \npointing to a pure tree state (with no history), or as in the case of git \nitself, there's a tag pointing to a blob that contains Junio's public key.\n\n> The situation is similar with the git tree:\n> \n> error: Object a0e7d36193b96f552073558acf5fcc1f10528917 is a blob, not a commit\n\nYeah, same thing.\n\n> Is this something I should be worried about?\n\nNo, but if it still happens with a newer git, holler.\n\n\t\t\tLinus\n"},{"id":"56437","messageId":"598D5675D34BE349929AF5EDE9B03E270168517F@az33exm24.fsl.freescale.net","threadId":"10322","inReplyTo":"alpine.LFD.0.999.0710181518120.26902@woody.linux-foundation.org","subject":"RE: linux-2.6.git mirror","fromName":"Medve Emilian-EMMEDVE1","fromEmail":"emilian.medve@freescale.com","sentAt":"2007-10-18T22:46:10Z","receivedAt":"2007-10-18T22:46:10Z","isPatch":false,"sender":{"key":"emilian.medve@freescale.com","avatar":null},"body":"Hi Linus,\n\n\n> > Is this something I should be worried about?\n> \n> No, but if it still happens with a newer git, holler.\n\nI tested this with Junio's latest master and a couple of stable releases\nfrom the maint branch with the same result. In my ignorance I suspected\nthe build infrastructure so I tried different gcc versions (4.x and 3.x)\nand optimization levels (including -O0) different libraries, etc. that I\nhad on a few machines around here. The result was the same.\n\nWhat worried me is that I think I traced the source of the error message\nin commit.c and in both two possible places from where the message could\ncome the processing flow seems to be cut shorter because of this.\n\n\nThanks for your reply,\nEmil.\n"},{"id":"56443","messageId":"alpine.LFD.0.999.0710181617090.26902@woody.linux-foundation.org","threadId":"10322","inReplyTo":"598D5675D34BE349929AF5EDE9B03E270168517F@az33exm24.fsl.freescale.net","subject":"RE: linux-2.6.git mirror","fromName":"Linus Torvalds","fromEmail":"torvalds@linux-foundation.org","sentAt":"2007-10-18T23:24:47Z","receivedAt":"2007-10-18T23:24:47Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Thu, 18 Oct 2007, Medve Emilian-EMMEDVE1 wrote:\n> \n> > > Is this something I should be worried about?\n> > \n> > No, but if it still happens with a newer git, holler.\n> \n> I tested this with Junio's latest master and a couple of stable releases\n> from the maint branch with the same result.\n\nOk, what is going on is:\n\n - append_fetch_head() looks up the SHA1 for all heads (including tags):\n\n if (get_sha1(head, sha1))\n return error(\"Not a valid object name: %s\", head);\n\n - it then wants to check if it's a candidate for merging (because \n fetching also does the whole \"list which heads to merge\" in case it is \n going to be part of a \"pull\"):\n\n commit = lookup_commit_reference(sha1);\n if (!commit)\n not_for_merge = 1;\n\n - and that \"lookup_commit_reference()\" is just very vocal about the case \n where it fails. It really shouldn't be, and it shouldn't affect the \n actual end result, but that basically explains why you get that scary \n warning.\n\nIn short, the warning is just bogus, and should be harmless, but I agree \nthat it's ugly. I think the appended patch should fix it.\n\nSigned-off-by: Linus Torvalds \n---\n\nAnd yes, I think this should go into Shawns tree of fixes, assuming that \nEmil confirms that it fixes it for him.\n\n\t\t\tLinus\n\n builtin-fetch--tool.c | 2 +-\n 1 files changed, 1 insertions(+), 1 deletions(-)\n\ndiff --git a/builtin-fetch--tool.c b/builtin-fetch--tool.c\nindex 1e43d79..e26817d 100644\n--- a/builtin-fetch--tool.c\n+++ b/builtin-fetch--tool.c\n@@ -131,7 +131,7 @@ static int append_fetch_head(FILE *fp,\n \n \tif (get_sha1(head, sha1))\n \t\treturn error(\"Not a valid object name: %s\", head);\n-\tcommit = lookup_commit_reference(sha1);\n+\tcommit = lookup_commit_reference_gently(sha1, 1);\n \tif (!commit)\n \t\tnot_for_merge = 1;\n \n"},{"id":"56641","messageId":"598D5675D34BE349929AF5EDE9B03E2701685385@az33exm24.fsl.freescale.net","threadId":"10322","inReplyTo":"alpine.LFD.0.999.0710181617090.26902@woody.linux-foundation.org","subject":"RE: linux-2.6.git mirror","fromName":"Medve Emilian-EMMEDVE1","fromEmail":"emilian.medve@freescale.com","sentAt":"2007-10-19T22:50:17Z","receivedAt":"2007-10-19T22:50:17Z","isPatch":false,"sender":{"key":"emilian.medve@freescale.com","avatar":null},"body":"Hi Linus,\n\n\n> And yes, I think this should go into Shawns tree of fixes, \n> assuming that \n> Emil confirms that it fixes it for him.\n\nIndeed, I don't get the error message anymore. Thanks for your help.\n\nA remaining question is why I wasn't seeing that error message on normal\nclones, i.e. non-mirrors (with +refs/heads/*:refs/remotes/origin/* fetch\nrefspec as oposed to +refs/*:refs/* fetch refspec)?\n\n\nThanks again,\nEmil.\n"}]}