Re: postgres: will not recognise existing tables

On 10/6/07, Darren Reed <darrenr(at)fastmail(dot)net> wrote:
> I've run into this problem twice in the last few days.
>
> postgresql is running, shutdown, reboot, restart, and tables are missing.
>
> The shutdown is clean:
> LOG: unexpected EOF on client connection
> LOG: received smart shutdown request
> LOG: shutting down
> LOG: database system is shut down
> LOG: database system was shut down at 2007-10-06 18:50:33 PDT
> LOG: checkpoint record is at 2/3310F20C
> LOG: redo record is at 2/3310F20C; undo record is at 0/0; shutdown TRUE
> LOG: next transaction ID: 228741; next OID: 16449
> LOG: next MultiXactId: 1; next MultiXactOffset: 0
> LOG: database system is ready
> LOG: transaction ID wrap limit is 2147484146, limited by database
> "postgres"
> LOG: received smart shutdown request
> LOG: shutting down
> LOG: database system is shut down
>
> The startup is clean:
> FATAL: lock file "postmaster.pid" already exists
> HINT: Is another postmaster (PID 99) running in data directory
> "/home/postgres/
> db"?

That's not a clean startup. The postmaster.pid should not exist.

Are sure you are not running postgresql in > 1 location on this machine...

Do you have an exact copy of a session where you see this happen?
I.e. the pg_ctl start, table updates, pg_ctl stop, pg_ctl start, table
corrupted? Including log files... I feel like we're kind of dancing
around the problem in the dark without enough info.

Please note this is absolutely non-normal behaviour. I've been using
pgsql since v6.5.2 and never had this occur to me.

As for your ulimit settings, what are they, specifically for the
postgres account on this machine, both explicit and inherited?