[ snipped description of setup ]
> > The 'ledgersmb.conf' file usually contains the name of the host to be
> > used as the database IP connection host. If you're using a database on
> > the local host, it's best to enter 'localhost' there - not the name of
> > the host in the DNS.
>
> It is set to 'localhost'.
>
Ok. You could try setting it to the path your unix sockets for postgresql
reside in, that allows bypassing the firewall. If nothing else, that's a
performance gain, most likely.
> > Actually, what I do when I run against a database which is local, I
> > enter the unix domain socket file's directory as the host name. On my
> > debian system that's /var/run/postgresql/. This reduces chances of your
> > firewall interfering with your database connection.
> >
> > The only errors I can find is this and seems related to login:
> >
> > /var/log/postgresql/postgresql-9.1-main.log:
> > 2013-07-14 16:53:58 CEST LOG: could not receive data from client:
> > Connection reset by peer
> >
> > /var/log/apache2/error.log:
> > [Sun Jul 14 16:53:58 2013] [error] [client 192.168.1.1] DBI
> > connect('dbname=nix','',...) failed: fe_sendauth: no password
> supplied
> > at LedgerSMB.pm line 986, referer:
> http://ledgersmb.kingel.net/login.pl
> >
> > Hmm. if you're able to log in afterwards, this is a bit unexplicable.
> > Does this happen on every page? or just when logging in?
>
> Just on login.
Ok. Then it can't explain the long response times on the later pages.
> > I have this in /usr/share/postgresql/9.1/pg_hba.conf:
> > host all all 127.0.0.1/32 http://127.0.0.1/32 md5
> > host all all 192.168.1.0/24 http://192.168.1.0/24 md5
>
Actually, my (8.4) pg_hba.conf lines are:
# "local" is for Unix domain socket connections only
local all all md5
# IPv4 local connections:
host all all 127.0.0.1/32 md5
# IPv6 local connections:
host all all ::1/128 md5
Note that mine don't list the second IP address that yours do. Maybe that's
part of the issue? (I have another instance with 9.2 on it, which looks
exactly like this one).
What do these two lines in pg_hba.conf mean?:
>
> @remove-line-for-nolocal@# "local" is for Unix domain socket connections
> only
> @remove-line-for-nolocal@local all all @authmethodlocal@
These lines look like they should not have been there. The "local" item is
the configuration for the "directory based" access method through Unix
sockets, but these lines look like remnants of a partially processed
template.
> > Ok. I assume you have restarted the postgresql server after you changed
> > its configuration files (such as pg_hba.conf)?
>
> Yes, more than once. :-)
Ok. Well, it's a bit mystical to me so far. Do the postgresql logs hold
information which might indicate problems? Maybe multiple rejected requests
in short succession?
--
Bye,
Erik.
http://efficito.com -- Hosted accounting and ERP.
Robust and Flexible. No vendor lock-in.
------------------------------------------------------------------------------
See everything from the browser to the database with AppDynamics
Get end-to-end visibility with application monitoring from AppDynamics
Isolate bottlenecks and diagnose root cause in seconds.
Start your free trial of AppDynamics Pro today!
http://pubads.g.doubleclick.net/gampad/clk?id=48808831&iu=/4140/ostg.clktrk
_______________________________________________
Ledger-smb-users mailing list
Ledger-smb-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ledger-smb-users
