Re: check for missing tablespaces?

* Andrew Dunstan (andrew(at)dunslane(dot)net) wrote:
> I am in the middle of helping a customer recover from a situation where
> a tablespace was missing when a machine was rebooted and postgres
> restarted, and I'm wondering if we should not have some sort of check
> for this on startup. Maybe we could check for the existence of the
> PG_VERSION file or something like that? Of course, that wouldn't help if
> the tablespace were subsequently lost, but it doesn't seem like a
> terribly expensive or unreasonable sanity check for startup.

I agree entirely with doing this. Now that we've got tablespaces and
our users are using them to split things on to seperate partitions which
could be accessed in different ways (eg: a tablespace on local storage
and one on a SAN) it makes alot of sense to make sure everything is in
order before allowing the system to completely come up. I'd much rather
PG not start when the SAN mappings aren't in place.

Thanks,

Stephen