Page Checksums

Folks,

What:

    Please find attached a patch for 9.2-to-be which implements page
    checksums.  It changes the page format, so it's an initdb-forcing
    change.

How:
    In order to ensure that the checksum actually matches the hint
    bits, this makes a copy of the page, calculates the checksum, then
    sends the checksum and copy to the kernel, which handles sending
    it the rest of the way to persistent storage.

Why:
    My employer, VMware, thinks it's a good thing, and has dedicated
    engineering resources to it.  Lots of people's data is already in
    cosmic ray territory, and many others' data will be soon.  And
    it's a TODO :)

If this introduces new failure modes, please detail, and preferably
demonstrate, just what those new modes are.  As far as we've been able
to determine so far, it could expose on-disk corruption that wasn't
exposed before, but we see this as dealing with a previously
un-dealt-with failure rather than causing one.

Questions, comments and bug fixes are, of course, welcome.

Let the flames begin!

Cheers,
David.
--
David Fetter <david@fetter.org> http://fetter.org/
Phone: +1 415 235 3778  AIM: dfetter666  Yahoo!: dfetter
Skype: davidfetter      XMPP: david.fetter@gmail.com
iCal: webcal://www.tripit.com/feed/ical/people/david74/tripit.ics

Remember to vote!
Consider donating to Postgres: http://www.postgresql.org/about/donate