Re: [PATCH 0/5] Number truncation with 4+ GB files on 32-bit systems

[Duy Nguyen <pclouds@gmail.com>]

On Wed, Jul 6, 2016 at 8:02 PM, Christoph Michelbach
<michelbach94@gmail.com> wrote:
> I now created a repo on a 64 bit system (same command as before), then
> duplicated it. One copy I applied git gc to on the 64 bit host system,
> the other copy I gave to my 32 bit virtual machine to apply git gc to
> it.
>
> The 64 bit host uses git from the Ubuntu repositories, the 32 bit
> virtual machine uses git 2.9 from github with the patches applied.
>
> git gc worked without problems on the host but I got
>
> frank@frank-virtual-16-04-32-bit:~/g$ git gc
> Counting objects: 6, done.
> Compressing objects: 100% (3/3), done.
> error: bad packed object CRC for
> f595ad71c1a1ecc312ddcb32a84a4bfc4a2ed1c8
> Writing objects: 100% (6/6), done.
> Total 6 (delta 0), reused 0 (delta 0)
> error: failed to validate delta base reference at offset 342896 from
> .git/objects/pack/pack-630b5a3f28cd9d02a546462bf0d0bc640ffde784.pack

Yup. So far I can only say that pack-objects generates incorrect pack
:-(  I hope it's the possible 6/5 patch I mentioned earlier but I'm
not sure. Will continue tomorrow.

> error: bad offset for revindex
> error: failed to read object 4246d27f8e0149d45687b0cc23bc29a67f1f0c79
> at offset 342887 from .git/objects/pack/pack-
> 630b5a3f28cd9d02a546462bf0d0bc640ffde784.pack
> fatal: packed object 4246d27f8e0149d45687b0cc23bc29a67f1f0c79 (stored
> in .git/objects/pack/pack-
> 630b5a3f28cd9d02a546462bf0d0bc640ffde784.pack) is corrupt
> error: failed to run prune
> frank@frank-virtual-16-04-32-bit:~/g$
>
> on the virtual machine.
>
> Not including the mailing list in CC wasn't intended.
>
> --
> With kind regards
> Christoph Michelbach
>
> On Wed, 2016-07-06 at 17:23 +0200, Duy Nguyen wrote:
>> On Wed, Jul 6, 2016 at 12:14 AM, Christoph Michelbach
>> <michelbach94@gmail.com> wrote:
>> >
>> > I now tried git gc again and it failed (in a different way, though;
>> > and
>> > the error message only appeared once git gc terminated).
>> >
>> > Full input and output:
>> >
>> > christoph@virt-16-04-32-bit:~$ mkdir test && cd test && git init &&
>> > touch someFile && git add someFile && git commit -m "Initial
>> > commit."
>> > && dd if=/dev/urandom of=bigBinaryFile bs=1MB count=4294 && git add
>> > bigBinaryFile && git commit -m "Introduced big biary file."
>> > Initialized empty Git repository in /home/christoph/test/.git/
>> > [master (root-commit) 20507ef] Initial commit.
>> >  1 file changed, 0 insertions(+), 0 deletions(-)
>> >  create mode 100644 someFile
>> > 4294+0 records in
>> > 4294+0 records out
>> > 4294000000 bytes (4.3 GB, 4.0 GiB) copied, 435.236 s, 9.9 MB/s
>> > [master 88e5dbb] Introduced big biary file.
>> >  1 file changed, 0 insertions(+), 0 deletions(-)
>> >  create mode 100644 bigBinaryFile
>> > christoph@virt-16-04-32-bit:~/test$ git gc
>> > Counting objects: 6, done.
>> > Compressing objects: 100% (3/3), done.
>> > Writing objects: 100% (6/6), done.
>> > Total 6 (delta 0), reused 1 (delta 0)
>> > error: inflate: data stream error (incorrect header check)
>> > error: failed to read object
>> > 705f438ccb845871ffba9d4b56f16ac763652937
>> Sigh.. I'll try again :) BTW if you have a 64-bit machine too, try
>> create the repo with it (so we are sure the repo is not corrupted in
>> the first place) then you can run tests like this on 32-bit systems.
>> The previous "bad crc" message was a fault in the code, not the data.
>>
>> PS. I have just let gc run till the end (last time I tried
>> pack-objects, which is only part of gc) and got a(nother) error.
>> Checking... BTW, you should send these mails to git@ as well, no need
>> for private message exchange.



-- 
Duy