Errors on a file on a zpool: How to remove?

[Greg 'groggy' Lehey]

[sequence recovered]

On Saturday, 23 January 2010 at 19:38:47 -0500, Rich wrote:
> On Sat, Jan 23, 2010 at 7:30 PM, Andrew Snow <andrew at modulus.org> wrote:
>> Rich wrote:
>>>
>>> I claim this is still Bad Behavior, and should be resolvable without
>>> doing something like that.
>>
>> I cannot agree that silent corruption (which would have happened with any
>> other filesystem) is preferable to what ZFS is doing here.
>>
>> You had bad RAM, and no redundancy in a huge RAID0, I think it would be
>> reasonable to have to restore from backups or recreate the data and not pin
>> blame on ZFS.
>
> I'm not claiming that restoring data from backups is unreasonable, or
> that silent data corruption is better.
>
> I'm claiming that my inability to delete the corrupted data in order
> to restore from backups without nuking unaffected data or remaking the
> filesystem is unreasonable.

You have a file system with broken metadata.  How can ZFS know how to
work around this corruption?  By continuing things could get worse.

I haven't been following this thread, but the way I see it:

Reasonable: to be able to mount the file system read-only and recover
what data you can from it.

Unreasonable: to expect the file system to repair corruption of an
unknown type.  The only safe path is to create the file system from
scratch.  This has nothing to do with how you recover the data.

Greg
--
See complete headers for address and phone numbers.
This message is digitally signed.  If your Microsoft MUA reports
problems, please read http://tinyurl.com/broken-mua
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 195 bytes
Desc: not available
Url : http://lists.freebsd.org/pipermail/freebsd-fs/attachments/20100124/45c46789/attachment.pgp