From owner-freebsd-bugs  Fri Sep 12 06:41:35 1997
Return-Path: <owner-freebsd-bugs>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.7/8.8.7) id GAA03502
          for bugs-outgoing; Fri, 12 Sep 1997 06:41:35 -0700 (PDT)
Received: from word.smith.net.au (word.smith.net.au [202.0.75.3])
          by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id GAA03488;
          Fri, 12 Sep 1997 06:41:30 -0700 (PDT)
Received: from word.smith.net.au (localhost.smith.net.au [127.0.0.1])
	by word.smith.net.au (8.8.7/8.8.5) with ESMTP id XAA03107;
	Fri, 12 Sep 1997 23:09:20 +1000 (EST)
Message-Id: <199709121309.XAA03107@word.smith.net.au>
X-Mailer: exmh version 2.0zeta 7/24/97
To: Graham Wheeler <gram@cdsec.com>
cc: mike@smith.net.au (Mike Smith), hackers@freebsd.org,
        freebsd-bugs@freebsd.org
Subject: Re: Memory leak in getservbyXXX? 
In-reply-to: Your message of "Sat, 12 Sep 1997 15:02:30 +0200."
             <199709121302.PAA20383@cdsec.com> 
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Fri, 12 Sep 1997 22:39:19 +0930
From: Mike Smith <mike@smith.net.au>
Sender: owner-freebsd-bugs@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

> 
> If the problem is a memory leak in the stdio stuff, then it seems a 
> reasonable assumption that if I do a setservent(1) at the start, that
> the problem will be circumvented, do you agree?

Presuming that the bug has to do with the open/close allocation, that 
sounds pretty reasonable to me.

Forgive me my confusion, but the following fragment here (I thought I 
should try to reproduce your leak) doesn't appear to behave as I would 
expect :

#include <netdb.h>
#include <stdio.h>

void main(void)
{
	for (;;)
		if (getservbyport(23, NULL) == NULL) {
			fprintf(stderr, "argh!\n");
			exit(1);
		}
}

ie. it exits immediately.  I tried various assortments of parameters 
(eg. network and native order on the port, NULL, "tcp", "udp", etc for 
the protocol).   8(

mike