I'm running PostgreSQL 7.1.
A few times now I have run out of disk space on the drive containing the postgresql data.
In all cases, this has corrupted the database, the postmaster has died, and refuses to start again
unless I rebuild from scratch.
As my app is under development at the moment, this isn't a big disaster, but I'm worrying
about what will happen when I go live.
The situation isn't helped by the postgres home directory being on /var, which could easily run out
of space for a number of reasons. (or is putting it in /var a Mandrake thing ? )
Obviously the solution is to put postgre on a disk with enough space, but I can see the App sitting
on the web server for a number of years until everyone forgets to check disk space - then BANG!
My question is:
Is it normal for postgres to get corrupted due to lack of disk space ?
Surely it should be able to handle this.
The problems have all occurred while processing a large transaction - could that be the cause ?
Thanks
JohnT