Subject: Re: [vserver] Possible Hashify Corruption
From: Gordan Bobic <gordan@bobich.net>
Date: Wed, 20 Oct 2010 20:01:07 +0100

On 20/10/2010 19:54, Michael S. Zick wrote:
> On Sun October 17 2010, Gordan Bobic wrote:
>> On 17/10/2010 14:54, Michael S. Zick wrote:
>>> On Sat October 16 2010, Gordan Bobic wrote:
>>>> Hi,
>>>>
>>>>
>>>> Can anybody hazard a guess as to what happened here? I'm prepared to
>>>> consider any theory at the moment, no matter how far fetched.
>>>>
>>>> I'm running 2.6.30.10-vs2.3.0.36.14-pre8. The file system is ext4
>>>> without journal and in data=writeback mode.
>>>>
>>>
>>> Lets go with your first guess, file corruption, and speculate a bit...
>>>
>>> We know that ext4 gets its speed by the high degree of meta-data and
>>> data catching that it uses.
>>> We know that if ext4 is not cleanly shut down, your file system is
>>> burnt toast.
>>> On any type of system.
>>
>> That is, in my experience, superstition.
>
> Well, your testing narrowed down that one -
>
> I must have been using one of the affected kernel versions when ext4
> scrambled 1Tbyte of files for me.  ;-)
>
> It was a copy of Fedora-11, but I don't recall which kernel version.

Fedora 11 sounds spot-on, actually. Fedora 12 came with 2.6.32, so F11 
would have been around the 2.6.30-2.6.31 mark, which right there with 
what I was using in this instance.

Gordan