[postgis-users] general postgreSQL

[Darren Houston]

Hello Eric,

Sorry for the late reply.

I had a deadlock problem with a JDBC application in Resin. Turns out that on
my RedHat 9 system, when I started X, after 2 days Resin would lock up. Only a
reboot would release   everything
(http://www.speculation.org/garrick/kill-9.html for a good quick tutorial on
killing processes). Kill -9 would have probably worked but I found reboot
enabled me to start PostgreSQL again (plus a reboot might be easier on a
system than a whole whack of kill -9's, but I could be wrong). I did a stack
trace (look up strace) and found out there was a call to a sleeping time
process that was not waking up so Resin couldn't write a timestamp to
PostgreSQL and hung the pool, or something like that. I am by no means an
expert in this area, so something I say could be wrong. I want to help you
because I know how frustrating it is when nobody has the answer and you don't
know how to go about solving the problem. A fix to my problem, don't start X.

Hope this helps some,

Darren H.


Eric.Hillmuth at anr.state.vt.us wrote:
>
> Thanks Darren;
>
> That postmaster.pid file cleans up at shutdown and port 5432 closes too. How should
> I go about looking for a connection pool deadlock?
>
> On 27 May 2004 at 18:28, Darren Houston wrote:
>
> > Hello Eric,
> >
> > One thought;  If PostgreSQL crashed and postmaster.pid is residing in your
> > PostgreSQL data directory, this might give the same error (just speculation).
> > With postmaster off there should be no postmaster.pid. If this is the case and
> > postmaster is off, delete postmaster.pid and try starting PostgreSQL again.
> > Also, I have had connection pools deadlock that seem to keep a perpetual
> > strangle hold on PostgreSQL, disabling a clean shutdown/restart.
> >
> > HTH somewhat,
> >
> > Darren H.
> >
> >
> > Eric.Hillmuth at anr.state.vt.us wrote:
> > >
> > > Its more related to PostgreSQL than PostGIS but I haven't gotten there yet... Has
> > > anyone run into an error like this from the postmaster on startup:
> > >
> > > LOG: could not bind IPv4 socket: Address already in use
> > > HINT: is another postmaster already running on port 5432. If not, wait a few
> seconds
> > > and retry.
> > >
> > > I can confirm that port 5432 is opened by the postmaster at startup and closed at
> > > shutdown with netstat etc. but I don't know why it prints this error. I'm running
> > > PostgreSQL 7.4 on SUSE 8.4.
> > >
> > > Any thoughts are appreciated.
> > >
> > >
> > >
> > >
> > >
> > > _______________________________________________
> > > postgis-users mailing list
> > > postgis-users at postgis.refractions.net
> > > http://postgis.refractions.net/mailman/listinfo/postgis-users
> > >
> >
> > _______________________________________________
> > postgis-users mailing list
> > postgis-users at postgis.refractions.net
> > http://postgis.refractions.net/mailman/listinfo/postgis-users
> >
>
>
>
> _______________________________________________
> postgis-users mailing list
> postgis-users at postgis.refractions.net
> http://postgis.refractions.net/mailman/listinfo/postgis-users
>