On Fri, Aug 17, 2007 at 09:59:26AM -0400, Tom Lane wrote:
> "Mikko Partio" <mpartio@gmail.com> writes:
> > This was my original intention. I'm still quite hesitant to trust the
> > fencing devices ability to quarantee that only one postmaster at a time is
> > running, because of the disastrous possibility of corrupting the whole
> > database.
>
> Making that guarantee is a fencing device's only excuse for existence.
> So I think you should trust that a properly-implemented fence will do
> what it's claimed to do.
>
> On the other side of the coin, I have little confidence in DRBD
> providing the storage semantics we need (in particular guaranteeing
> write ordering). So that path doesn't sound exactly risk-free either.
>
> regards, tom lane
of course we guarantee write ordering.
we (linbit, company behind drbd, paying drbd developers)
operate quite a few postgres clusters in production on clusters
"powered by heartbeat and DRBD".
there are much more we do not operate directly ourselves.
just because we happen to have a partnership with mysql
does not mean we don't like postgres very much indeed :)
to get an idea of what drbd does for you, please,
if you are interessted, read some of the
http://www.drbd.org/publications.html,
maybe start with the LinuxConf 2007 pdf.
cheers,
--
: Lars Ellenberg Tel +43-1-8178292-55 :
: LINBIT Information Technologies GmbH Fax +43-1-8178292-82 :
: Vivenotgasse 48, A-1120 Vienna/Europe http://www.linbit.com :
