Hi,
In the past few days I've continued to examine some
replication possibilities for PostgreSQL, in order to
build a high-availability setup, but I haven't found
what I've been looking for.
So, now I'm considering a different approach to high-
availability, without using replication. I'm not sure
if this is really feasible or not ... Let me explain.
In our particular case, all of the storage (including
database files) is mounted from a NetApp filer via NFS.
Actually it's a cluster consisting of two Filers which
are mirrored, and each of them uses RAID-5, so the
storage is already redundant and highly available.
So my idea was not to replicate, but to use a second
machine as a hot-standby server which would mount the
very same database files via NFS. It could already be
booted and have everything mounted, but the PostgreSQL
daemon is not running. As soon as a failure of the
first machine is detected, some kind of mechanism would
startup the daemon on the hot-standby machine.
I am aware that two PostgreSQL servers must NEVER access
the same database files at the same time. Assuming that
I can guarantee that -- would the solution work?
Or even better: Has anyone done something like that
before? Maybe there are already some ready-made scripts?
Any hint or advice would be greatly appreciated.
Thankyou very much in advance!
Best regards
Oliver
--
Oliver Fromme, secnetix GmbH & Co KG, Oettingenstr. 2, 80538 München
Any opinions expressed in this message may be personal to the author
and may not necessarily reflect the opinions of secnetix in any way.
"Unix gives you just enough rope to hang yourself --
and then a couple of more feet, just to be sure."
-- Eric Allman
