[OmniOS-discuss] (no subject)

[Paul B. Henson]

> From: Stephan Budach
> Sent: Monday, September 14, 2015 10:00 PM
>
> As George Wilson wrote on the ZFS mailing list: " Unfortunately, if the
> corruption impacts a data block then we won't be able to detect it.".
> So, I am afarid apart from metadata and indirect blocks corruption,
> there's no way to even detect a corruption inside a data block, as the
> checksum fits.

Yes, that's true, assuming you have no external source of verification.
However, Arne said he didn't think this bug would result in data corruption,
only metadata corruption. I was mostly worried about pool corruption that
would cause panics or failure to import, which data level corruption would
not cause. Most of the data on the pool I was worried about is media, a bad
data block here or there wouldn't be too tragic.

> from that pool, e.g. from a backup prior to 6214 having been introduced,
> but depending on the sheer amount of data or the type of it, that might
> not be even possible.

Yup. This was a sucky bug :(.