Message-ID: <4D62E236.8000605@cornell.edu>
Date: Mon, 21 Feb 2011 17:07:50 -0500
From: Ken Brown <kbrown@cornell.edu>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US; rv:1.9.2.13) Gecko/20101207 Thunderbird/3.1.7
MIME-Version: 1.0
To: cygwin@cygwin.com
Subject: Re: Spurious "Connection reset by peer" problems anybody?
References: <20110221104352.GA17868@calimero.vinschen.de> <4D626D65.3060408@cornell.edu> <20110221140140.GD17868@calimero.vinschen.de> <20110221142606.GE17868@calimero.vinschen.de> <4D627C96.8000807@cornell.edu> <20110221150636.GF17868@calimero.vinschen.de> <4D62843F.2000205@cornell.edu> <20110221153406.GA11901@calimero.vinschen.de>
In-Reply-To: <20110221153406.GA11901@calimero.vinschen.de>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Mailing-List: contact cygwin-help@cygwin.com; run by ezmlm
Sender: cygwin-owner@cygwin.com
Mail-Followup-To: cygwin@cygwin.com

On 2/21/2011 10:34 AM, Corinna Vinschen wrote:
> On Feb 21 10:26, Ken Brown wrote:
>> On 2/21/2011 10:06 AM, Corinna Vinschen wrote:
>>> As far as this problem goes, it would only occur at the end of a
>>> connection, when close() is called on the socket on the Windows side.
>>> The error message would occur on the peer side, independent of the OS
>>> it's running on.  If the connection just breaks at some arbitary point,
>>> it's almost surely not the problem I'm talking about.
>>
>> In my case the connection does break at some point in the middle,
>> not at the end, so it sounds like I have a different problem.
>
> Given that this is a bzr+ssh scenario, there's a good chance that your
> problem is fixed by the patch cgf applied tonight.  Would you mind to
> test this with the latest snapshot?

The problem is not fixed, but I think it's happening less often.  (I 
can't be sure, since the problem is sporadic.)  Here's what I did:

1. I installed the snapshot and updated my mirror of the bzr repository. 
  It failed once but then worked the second time.

2. I tried (once) to create a new mirror from scratch, and it failed.

3. I rebuilt cygwin1.dll with your latest changes (2011-02-21 17:01:06) 
and tried again to create a new mirror from scratch.  Again it failed 
once but worked the second time.  This is remarkable, because a huge 
amount of data got transferred.  So I'm inclined to think there's been 
an improvement.  But I might have just gotten lucky.

Ken

--
Problem reports:       http://cygwin.com/problems.html
FAQ:                   http://cygwin.com/faq/
Documentation:         http://cygwin.com/docs.html
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple