Mailing-List: contact cygwin-developers-help AT cygwin DOT com; run by ezmlm
List-Subscribe: <mailto:cygwin-developers-subscribe AT cygwin DOT com>
List-Archive: <http://sources.redhat.com/ml/cygwin-developers/>
List-Post: <mailto:cygwin-developers AT cygwin DOT com>
List-Help: <mailto:cygwin-developers-help AT cygwin DOT com>, <http://sources.redhat.com/ml/#faqs>
Sender: cygwin-developers-owner AT cygwin DOT com
Delivered-To: mailing list cygwin-developers AT cygwin DOT com
Message-ID: <038501c2df54$8797e1c0$78d96f83@pomello>
From: "Max Bowsher" <maxb AT ukf DOT net>
To: <cygwin-developers AT cygwin DOT com>
References: <20030228174454 DOT GA11125 AT redhat DOT com> <20030228175114 DOT GO20955 AT cygbert DOT vinschen DOT de> <20030228175455 DOT GP20955 AT cygbert DOT vinschen DOT de>
Subject: Re: 1.3.21 showstopper AF_UNIX problems?
Date: Fri, 28 Feb 2003 18:09:26 -0000
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-Priority: 3
X-MSMail-Priority: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1106

Corinna Vinschen wrote:
> On Fri, Feb 28, 2003 at 06:51:14PM +0100, Corinna Vinschen wrote:
>> On Fri, Feb 28, 2003 at 12:44:54PM -0500, Christopher Faylor wrote:
>>> Two people have now reported AF_UNIX problems.  I can't duplicate
>>> them but, then, there wasn't many details provided.  I wouldn't
>>> expect them from the guy who runs the perl test suite, though.
>>>
>>> Can anyone here duplicate any problems?
>>
>> Besides the perl guy, did I miss #2?  Or are you talking about the
>> guy with the fstat vs. stat problem on AF_UNIX sockets?
>
> Uh, what about Max Bowsher mentioning a socketpair(AF_UNIX) problem
> without giving any details?  Max?  What's up?  Where's your testcase?

The only socketpair problem I mentioned involved TCP sockets. I haven't
nailed it down, but it seems to have something to do with exhausting
available addresses - the only times I've been able to cause it to happen
are with >150 connections in TIME_WAIT. I.e. no a showstopper.

Max.