Re: [INTERFACES] SLOWNESS ON JDBC INTERFACE

Tks for the clue ..... I had a bogus resolv.conf file and
it wasn't doing lookups correctly.

Appreciate the help.

Mike Meyer


----------
> From: Peter T Mount <peter@retep.org.uk>
> To: Debbie and Mike Meyer <mwmeyer@omeganet.es>
> Cc: pgsql-interfaces@postgreSQL.org
> Subject: Re: [INTERFACES] SLOWNESS ON JDBC INTERFACE
> Date: Saturday, September 26, 1998 12:32 PM
>
> On Fri, 25 Sep 1998, Debbie and Mike Meyer wrote:
>
> > I've been installing and learning the jdbc driver with 6.3.2.
> > The example programs in the ../postgresql/src/interface/jdbc
> > directory run fine -- except everyone takes a good two
> > minutes to execute.
> >
> > Running in debug, the hangup is on this line:
> >
> > [Loaded java/net/InetAddressImpl.class from
> > /usr/lib/jdk-1.1.5/lib/classes.zip]
> >
> > It halts there for a couple of minutes then makes the connection
> > quickly.
> >
> > It's jdk-1.1.5 from the Redhat RPM.  Redhat 5.0 upgraded
> > to kernel 2.0.35.  postmaster is running with tcpip ( -i option)
> > The connection is from localhost (127.0.0.1) and the pg_hba.conf is
> > left at default (with localhost permissions wide open).
>
> I assume you mean the "host all 127.0.0.1" line is uncommented...
>
> > It's only a P-75 on a hobby machine, but I think it shud
> > be more responsive.  Normal commands from within
> > psql work fine.
>
> Even then it should be more responsive than that.
>
> >
> > Anyone seen this problem or at least have some ideas
> > how to approach troubleshooting it?
>
> All of the examples are written on a P133 Linux box, and they show no
> problems of this nature.
>
> I have seen this symptom on other systems (nothing to do with postgresql
> or java), and it's either been caused by a routing or host lookup
failure.
>
> For example: At work, I have a large corporate unix machine. If a PC is
> not in that machines /etc/hosts file, then there is a delay of 60-120
> seconds before a connection is made.
>
> PS: JDK 1.1.5 is ok for pre v6.4 versions of the driver. In 1.1.6, Sun
> fixed a bug in the java.util.Date class, and v6.4 now handles dates how
> they should be, so it can break on 1.1.5 or earlier. The earlier drivers
> had to add a day when storing them in PreparedStatement.setDate()
>
> Peter
>
> --
>        Peter T Mount peter@retep.org.uk
>       Main Homepage: http://www.retep.org.uk
> PostgreSQL JDBC Faq: http://www.retep.org.uk/postgres
>  Java PDF Generator: http://www.retep.org.uk/pdf