I'm sure it's using that pg_hba.conf, although I don't see anything in the
log file that indicates the file it is actually using. I shut the postmaster
down, modified it, and started the postmaster.
I tried using 'trust' and the clients are able to connect. Could it be a
case-sensitivity problem with the database name (seeing as it's all caps)?
-----Original Message-----
From: Tom Lane [mailto:tgl@sss.pgh.pa.us]
Sent: Monday, April 18, 2005 1:45 PM
To: Walker, Jed S
Cc: 'pgsql-novice@postgresql.org'
Subject: Re: [NOVICE] Authentication problem
"Walker, Jed S" <Jed_Walker@cable.comcast.com> writes:
> I have the postgresql.conf listen_addresses='*'
> and in pg_hba.conf I have
> host VRNJED all 2.30.0.0 255.255.0.0 md5
That looks fine ...
> I have three clients on Windows PCs each on a different subnet
> (2.30.50, 2.30.190, 2.30.31). One of them connects just fine, the
> password is asked for and the logon succeeds, but the other two get
> C:\dev\PostgreSQL>psql -h robux -d VRNJED -U vrngui
> psql: FATAL: password authentication failed for user "vrngui"
> I'm not sure why these two fail when the others work.
Odd. One possibility is that the postmaster isn't using the pg_hba.conf
file you think it is (did you signal it to reload its configuration, or else
stop and restart it, after editing the file?). Another idea that comes to
mind is some sort of packet filtering at the network or OS level ... though
I'm not sure how packet filtering could cause this particular error message.
regards, tom lane