Re: PG on zeta 1.0

2006-05-20 (토), 12:39 -0400, Tom Lane 쓰시길:
> kmh496 <kmh496@kornet.net> writes:
> > 2006-05-20 (), 11:53 -0400, Tom Lane ñ:
> >> It looks to me like you don't have DNS correctly configured on that
> >> machine.  Every machine should be able to resolve "localhost" as
> >> "127.0.0.1" --- that's standard.  Look at name resolution settings,
> >> /etc/resolv.conf, that sort of thing.
>
> > i think it's the port for an unknown reason.  I don't have any port
> > number listed for 5543 but it gives me the same error when i try to use
> > it.  If I use port 55543 however there is no problem.
>
> Really?  That's very interesting.  Have you tried adding 5432 to
> /etc/services and then referencing the port number by the name you
> give it there?  I've never heard of a resolver rejecting numerical
> specifications for low-numbered ports, but there's a first time for
> everything ...
>
mr. lane, you are correct.  as it turns out it is the combination of
"-h/mnt/bu/mod -p5543".... actually, it's just the -h part.... somehow i
thought it would take a directory for the hostname.

the problem becomes that it's started and running but you can't connect
to it through the socket (locally), so it's confusing what's going on.

[postgres@www ~]$ cat /etc/services | grep post
pop2            109/tcp         pop-2   postoffice      # POP version 2
postgres        5432/tcp                        # POSTGRES
postgres        5432/udp                        # POSTGRES
[postgres@www ~]$
[postgres@www ~]$ pg_ctl -D pg8 -o "-h/mnt/bu/pg8" start
postmaster starting
[postgres@www ~]$ LOG:  could not translate host name "/mnt/bu/pg8",
service "5432" to address: Name or service not known
WARNING:  could not create listen socket for "/mnt/bu/pg8"
LOG:  database system was shut down at 2006-05-21 01:57:47 KST
LOG:  checkpoint record is at 1/290936C
LOG:  redo record is at 1/290936C; undo record is at 0/0; shutdown TRUE
LOG:  next transaction ID: 7315307; next OID: 11726712
LOG:  database system is ready
[postgres@www ~]$

p.s.  i hope the problem with the encoding (see other email ) was
connecting through the wrong port to the  mod database in another
cluster.