On Sun, Apr 24, 2011 at 6:22 PM, Simon Riggs <simon@2ndquadrant.com> wrote:
> My implementation path for that would be to add a crash_number onto
> pg_control and pg_index. Any index marked as "unlogged, persistent"
> would only be usable if it's crash number is the same as current
> system crash number.
>
> REINDEX would update the index crash number to current value. That
> also allows us to imagine a "repair index" command in the future as
> well.
This seems useful for non-crash-safe indexes in general.
> Heap blocks would be zeroed if they were found to be damaged, following a crash.
>
How do you propose to detect that? Until we solve the whole checksum
story I don't think we have a reliable way to detect bad pages. And in
some cases where do detect them we would detect them by crashing.
--
greg