Message-ID: <37642917.FED11C6B@uibk.ac.at>
Date: Sun, 13 Jun 1999 23:56:39 +0200
From: Georg Ritter <Georg.Ritter@uibk.ac.at>
MIME-Version: 1.0
To: pgsql-sql@hub.org, pgsql-interfaces@hub.org
Subject: backing up a db, jdbc, SQL syntax, etc...
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Hello together!

I come up with a conglomerat of questions ;-)

o What do you think is the best way to backup the data stored in a 
  postgres db? Backing up the whole db-file, doing a dump into ascii files with
  seperators or whater ever method else?


o Is there a maximum number of connections to the backend that could 
  run us into trouble if lot's of users who are connected via jdbc do not
  close their applets? 

o Do I have to take any speacial locking mechanism in SQL into account 
  for a high number (~500) of users. I only use SELECTs, UPDATEs, INSERTs and 
  NEXTVALUE from a sequence over jdbc. Problems with caches, memory/connection
   ...?

o Does anyone know how I can throw out (stop) a java applet completly out of the 
  java interpreter in a browser; so that all resources and memory are released.
  In the moment this only happens when I kill the browser (netscape on linux)

  Here some Q's which I posted some times ago in the novice-list but where 
  not answered and so are still open (for me, naturally :-) )

o Where is the difference between:

        SELECT f.title, f.did, d.name FROM dist d, films f WHERE f.did=d.did;
  and
        SELECT films.title, films.did, dist.name FROM dist JOIN films
        USING(did);

  or a formulation with inner join? Don't the result in the same output? 

  If I'm not mistaken, why do I get a inner join not yet implemented
  information, would a inner join have to be implemented in a different 
  way in the db-backend?


Greetings from Austria, 

        Georg Ritter
 

-- 
---------------------------------------------------------------------
  student at department of physics
  University of Innsbruck, Austria            Georg.Ritter@uibk.ac.at