ZFS under FreeBSD failure modes

[Andrew Berg]

On 2014.08.05 20:19, Paul Kraus wrote:
> On Aug 2, 2014, at 17:38, kpneal at pobox.com wrote:
> 
>> I'd be careful running ZFS on a machine that lacks ECC memory. Lots of
>> people do it, but I'd be worried that ZFS would get itself into a state
>> where you couldn't access anything.
> 
> I am startring to see comments like this on a more frequent basis. What is the failure mechanism you expect to run into here?
The idea is that a bad block in RAM will get written to disk and ZFS will not
know it is bad and then create a checksum based on it and call it good.
However, UFS and other filesystems would be just as incapable of detecting and
correcting the error. Yes, ECC RAM is always better than non-ECC RAM, but ZFS
is still going to be better equipped to detect errors than UFS. Not having ECC
RAM is more reason to use ZFS, not less.

>> When was the last time anyone heard of a UFS file
>> system being so damaged that it couldn't be recovered?
> 
> Anecdotal evidence at best. I have plenty of anecdotal evidence that ZFS never looses data. I don’t claim it as fact. 
The thing to remember is that ZFS is far more capable of finding data
corruption than UFS and it will complain loudly whether it can correct the
situation or not. To the person unfamiliar with ZFS, this can make it seem like
ZFS causes more problems when in fact ZFS is just finding problems that were
already there that another filesystem would miss.