{"thread":{"id":"23772","subject":"Help with a possibly corrupt repo","startedAt":"2010-05-10T19:37:36Z","lastAt":"2010-05-21T02:59:54Z","messageCount":6,"participants":["Jonathan Leto","Dmitry Potapov"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"141431","messageId":"AANLkTiks5g7BvLXBrUcb0m5dCiIrBifjDUKBDScXd1dw@mail.gmail.com","threadId":"23772","inReplyTo":null,"subject":"Help with a possibly corrupt repo","fromName":"Jonathan Leto","fromEmail":"jaleto@gmail.com","sentAt":"2010-05-10T19:37:36Z","receivedAt":"2010-05-10T19:37:36Z","isPatch":false,"sender":{"key":"jaleto@gmail.com","avatar":"https://gravatar.com/avatar/d0cd58a65ab4a83968329c53be867890ca0d50c6e54ae7683fa2f61f347db221?d=mp&s=160"},"body":"Howdy,\n\nDue to a network (or possibly disk) error, I started getting this error:\n\n$ git --version\ngit version 1.7.1.166.gf2086\nerror: garbage at end of loose object '66fdbaef5ae12a8402f6b0f7c1a73a3ce72e8eba'\nfatal: object 66fdbaef5ae12a8402f6b0f7c1a73a3ce72e8eba is corrupte$ git reflog\n291cf66 HEAD@{0}: commit: Class::MethodMaker\n44c9cdb HEAD@{1}: commit: TAP::Harness::Archive\n89053c5 HEAD@{2}: HEAD^: updating HEAD\nb6eb92a HEAD@{3}: commit: TAP::Harness::Archive\n89053c5 HEAD@{4}: commit: Bio::Perl 1.006001\nc745fed HEAD@{5}: commit: MooseX::Declare\nd4097d4 HEAD@{6}: commit (initial): Initial .modulebuildrc\nerror: garbage at end of loose object '66fdbaef5ae12a8402f6b0f7c1a73a3ce72e8eba'\nfatal: object 66fdbaef5ae12a8402f6b0f7c1a73a3ce72e8eba is corrupted\n\n$ uname -a\nLinux datura 2.6.31-21-generic #59-Ubuntu SMP Wed Mar 24 07:28:27 UTC\n2010 x86_64 GNU/Linux\n\nThis git was built from the 'next' branch a few days ago. Is there\nanyway to fix my repo? I can provide a copy of my .git directory if\nneeded, but it is 54MB. As for background, this repo contained a set\nof CPAN modules, via local::lib, so it had many files but only a few\ncommits.\n\nI tried \"git gc\", but it borks on the corrupt sha1:\n\n$ git gc\nerror: garbage at end of loose object '66fdbaef5ae12a8402f6b0f7c1a73a3ce72e8eba'\nfatal: object 66fdbaef5ae12a8402f6b0f7c1a73a3ce72e8eba is corrupted\nerror: failed to run repack\n\nThanks for any help anybody can give.\n\nDuke\n\n\n-- \nJonathan \"Duke\" Leto\njonathan@leto.net\nhttp://leto.net\n"},{"id":"141433","messageId":"20100510201720.GH14069@dpotapov.dyndns.org","threadId":"23772","inReplyTo":"AANLkTiks5g7BvLXBrUcb0m5dCiIrBifjDUKBDScXd1dw@mail.gmail.com","subject":"Re: Help with a possibly corrupt repo","fromName":"Dmitry Potapov","fromEmail":"dpotapov@gmail.com","sentAt":"2010-05-10T20:17:20Z","receivedAt":"2010-05-10T20:17:20Z","isPatch":false,"sender":{"key":"dpotapov@gmail.com","avatar":"https://avatars.githubusercontent.com/u/6568595?v=4"},"body":"On Mon, May 10, 2010 at 12:37:36PM -0700, Jonathan Leto wrote:\n> Due to a network (or possibly disk) error, I started getting this error:\n> \n> $ git --version\n> git version 1.7.1.166.gf2086\n> error: garbage at end of loose object '66fdbaef5ae12a8402f6b0f7c1a73a3ce72e8eba'\n\nThe object is \"loose\", so it should be a file:\n\n.git/objects/66/fdbaef5ae12a8402f6b0f7c1a73a3ce72e8eba\n\nit would be interesting to look at it to see what is wrong...\n\n\nDmitry\n"},{"id":"141435","messageId":"AANLkTinYj624545WtRf8TayNkm6a_z3e6kiwRyHCjqtv@mail.gmail.com","threadId":"23772","inReplyTo":"20100510201720.GH14069@dpotapov.dyndns.org","subject":"Re: Help with a possibly corrupt repo","fromName":"Jonathan Leto","fromEmail":"jaleto@gmail.com","sentAt":"2010-05-10T21:05:09Z","receivedAt":"2010-05-10T21:05:09Z","isPatch":false,"sender":{"key":"jaleto@gmail.com","avatar":"https://gravatar.com/avatar/d0cd58a65ab4a83968329c53be867890ca0d50c6e54ae7683fa2f61f347db221?d=mp&s=160"},"body":"Howdy,\n\nI have attached the loose file, hopefully that will help. This may\nalso be useful:\n$ hexdump .git/objects/66/fdbaef5ae12a8402f6b0f7c1a73a3ce72e8eba|tail\n0002060 b379 5619 df70 0a6f 6bdf 570d 9a52 9257\n0002070 045a d552 68d8 bd31 ddaf 7bed 73b8 dcf1\n0002080 3dd9 f773 3874 824d e419 5891 019c 9252\n0002090 3efb fdc4 99f8 9f85 f697 926c f9bf 5cb1\n00020a0 ff7a eb8c 5055 8283 8a4b f332 41d2 ea8a\n00020b0 480f e52b 6e97 947c a2ea a65e 393b d98b\n00020c0 deed 3541 7800 56dd 005c 0000 0000 0000\n00020d0 0000 0000 0000 0000 0000 0000 0000 0000\n*\n0003000\nerror: garbage at end of loose object '66fdbaef5ae12a8402f6b0f7c1a73a3ce72e8eba'\nfatal: object 66fdbaef5ae12a8402f6b0f7c1a73a3ce72e8eba is corrupted\n\nDuke\n\n\n\n-- \nJonathan \"Duke\" Leto\njonathan@leto.net\nhttp://leto.net\n"},{"id":"141438","messageId":"20100510212801.GJ14069@dpotapov.dyndns.org","threadId":"23772","inReplyTo":"AANLkTinYj624545WtRf8TayNkm6a_z3e6kiwRyHCjqtv@mail.gmail.com","subject":"Re: Help with a possibly corrupt repo","fromName":"Dmitry Potapov","fromEmail":"dpotapov@gmail.com","sentAt":"2010-05-10T21:28:01Z","receivedAt":"2010-05-10T21:28:01Z","isPatch":false,"sender":{"key":"dpotapov@gmail.com","avatar":"https://avatars.githubusercontent.com/u/6568595?v=4"},"body":"On Mon, May 10, 2010 at 02:05:09PM -0700, Jonathan Leto wrote:\n> 00020c0 deed 3541 7800 56dd 005c 0000 0000 0000\n> 00020d0 0000 0000 0000 0000 0000 0000 0000 0000\n> *\n> 0003000\n> error: garbage at end of loose object '66fdbaef5ae12a8402f6b0f7c1a73a3ce72e8eba'\n> fatal: object 66fdbaef5ae12a8402f6b0f7c1a73a3ce72e8eba is corrupted\n\nThe object is \"tree\", and it contains a lot of zeros at end.\nI have truncated them, and it appears to fix the problem.\nI have attached the corrected file.\n\nBut the important question is how it could happen. What filesystem\ndo you use? Do you store it on the local disk or network share?\n\n\nDmitry\n\n\nx\u0001+)JMU024c040031Qp,*J��+�e��d�2h����[���Տ/q]\u0014�\nU㔟�\u0003R��k�}��k���3d\u0015~�xQyq�\"T�s~J*H�L�S7\u0007��V�}y�\u0019Vp�o\n�k\rWR�W�Z\u0004R��h1����{�s���=s�t8M�\u0019�X�\u0001R��>��������l����\\z���UP��K�2��A��\u000fH+�n|��^�;9����A5x�V\\"},{"id":"141967","messageId":"AANLkTikSo8hTPXG8kGgCLOul6wp8THobUW_rSni4D6YC@mail.gmail.com","threadId":"23772","inReplyTo":"20100510212801.GJ14069@dpotapov.dyndns.org","subject":"Re: Help with a possibly corrupt repo","fromName":"Jonathan Leto","fromEmail":"jaleto@gmail.com","sentAt":"2010-05-20T14:57:30Z","receivedAt":"2010-05-20T14:57:30Z","isPatch":false,"sender":{"key":"jaleto@gmail.com","avatar":"https://gravatar.com/avatar/d0cd58a65ab4a83968329c53be867890ca0d50c6e54ae7683fa2f61f347db221?d=mp&s=160"},"body":"Howdy,\n\nI ran into this corruption issue using ecryptfs with an ext4\nfilesystem (on a local SSD drive) on Ubuntu 9.10. The corruption seems\nto coincide with getting these errors in my dmesg output:\n\n[12773.772426] ecryptfs_decrypt_page: Error attempting to read lower\npage; rc = [-22]\n[12773.772431] ecryptfs_readpage: Error decrypting page; rc = [-22]\n[12796.061574] ecryptfs_read_lower: octets_read = [-4]; expected [4096]\n[12796.061584] ecryptfs_read_and_validate_header_region: Error reading\nheader region; rc = [-22]\n\nI am lead to believe that ecryptfs+ext4 is not yet production-ready,\nalthough it could be a weird failure mode of my drive. I have not had\nmore time to test this setup with other hardware.\n\nDuke\n\n\nOn Mon, May 10, 2010 at 2:28 PM, Dmitry Potapov <dpotapov@gmail.com> wrote:\n> On Mon, May 10, 2010 at 02:05:09PM -0700, Jonathan Leto wrote:\n>> 00020c0 deed 3541 7800 56dd 005c 0000 0000 0000\n>> 00020d0 0000 0000 0000 0000 0000 0000 0000 0000\n>> *\n>> 0003000\n>> error: garbage at end of loose object '66fdbaef5ae12a8402f6b0f7c1a73a3ce72e8eba'\n>> fatal: object 66fdbaef5ae12a8402f6b0f7c1a73a3ce72e8eba is corrupted\n>\n> The object is \"tree\", and it contains a lot of zeros at end.\n> I have truncated them, and it appears to fix the problem.\n> I have attached the corrected file.\n>\n> But the important question is how it could happen. What filesystem\n> do you use? Do you store it on the local disk or network share?\n>\n>\n> Dmitry\n>\n\n\n\n-- \nJonathan \"Duke\" Leto\njonathan@leto.net\nhttp://leto.net\n"},{"id":"142003","messageId":"20100521025954.GA3005@dpotapov.dyndns.org","threadId":"23772","inReplyTo":"AANLkTikSo8hTPXG8kGgCLOul6wp8THobUW_rSni4D6YC@mail.gmail.com","subject":"Re: Help with a possibly corrupt repo","fromName":"Dmitry Potapov","fromEmail":"dpotapov@gmail.com","sentAt":"2010-05-21T02:59:54Z","receivedAt":"2010-05-21T02:59:54Z","isPatch":false,"sender":{"key":"dpotapov@gmail.com","avatar":"https://avatars.githubusercontent.com/u/6568595?v=4"},"body":"On Thu, May 20, 2010 at 07:57:30AM -0700, Jonathan Leto wrote:\n> \n> I ran into this corruption issue using ecryptfs with an ext4\n> filesystem (on a local SSD drive) on Ubuntu 9.10.\n> to coincide with getting these errors in my dmesg output:\n> \n> [12773.772426] ecryptfs_decrypt_page: Error attempting to read lower\n> page; rc = [-22]\n> [12773.772431] ecryptfs_readpage: Error decrypting page; rc = [-22]\n> [12796.061574] ecryptfs_read_lower: octets_read = [-4]; expected [4096]\n> [12796.061584] ecryptfs_read_and_validate_header_region: Error reading\n> header region; rc = [-22]\n> \n> I am lead to believe that ecryptfs+ext4 is not yet production-ready,\n\nSounds like this bug:\nhttps://bugs.launchpad.net/ecryptfs/+bug/509180\nand maybe related to this:\nhttps://bugs.launchpad.net/ubuntu/+source/linux/+bug/490005\n\nI don't think it is caused by combination of ecryptfs+ext4. I have seen\na similar bug report for ecryptfs+ext3 as well. In fact, a lot of bug\nfixes were merged for ecryptfs during 2.6.34 development cycle.\nPerhaps, those fixes were also back ported to 2.6.32, but Ubuntu 9.10\nhas 2.6.31, so it does not have them, and without them, it looks like\na beta-quality filesystem. So, make sure you have regular backups!\n\n\nDmitry\n"}]}