[Swan] Before considering how Libreswan can be better coded

[Philippe Vouters]

Paul,

When I took into account your remark that my fix apparently broke your 
tests, you gave me the ones you usually ran and allowed me a remote 
access to your test machine (with cons north-east and south-west). As 
usual when I want to observe what's wrong, I ALWAYS run a test manually. 
The tests you designed and that I ran did not work but, as far as I 
could observe, the very reason why they failed was absolutely and 
obviously totally decorrelated to my fix you wanted to draw my attention 
onto. Dig into your mail history and reread the mails I sent you at that 
time.

Yours truly,

Philippe Vouters (Fontainebleau/France)
URL: http://vouters.dyndns.org/
SIP: sip:Vouters at sip.linphone.org

On 05/24/2013 08:10 PM, Paul Wouters wrote:
> On Fri, 24 May 2013, Philippe Vouters wrote:
>
>> Look first at your bug I cured many times instead of hurrying to 
>> criticize others.
>
> I and others looked at it and showed you it was the wrong fix for your
> problem. It caused significant damage and broke dozens of test cases.
>
>> This configuration worked with Openswan 2.6.38 and below and works 
>> with Libreswan 3.0 as well. Libreswan 3.1 and onward broke this 
>> configuration.
>
> Yes, I admit we broke your broken configuration that you keep refusing
> to fix. We will get to fixing bug #102, as I states before, hopefully
> for the 3.5 release.
>
> Paul
>