[ntp:questions] Reference implementation 4.2.4p6 receive() swapping bytes..?

regan_russell at hotmail.com regan_russell at hotmail.com
Tue Mar 10 05:05:57 UTC 2009


On Mar 10, 1:53 pm, ma... at ntp.org (Danny Mayer) wrote:
> regan_russ... at hotmail.com wrote:
> > Hi there,
>
> > In ntp_config.c I have added my own modifiers on the server line so
> > that if talking to one of my own servers I flag it as such, in the
> > peer structure I add a flag to indicate own of my category of time
> > servers, I added code to ntp_crypto.c and ntp_protocol.c and have my
> > own crypto type and shove my own stuff into extension fields. All the
> > stuff I am doing works but...
>
> Exactly what are you trying to accomplish? Obviously you are doing
> something wrong but it would be better if you explained why you are
> doing it so we can suggest better options. It is dangerous to muck
> around with the ntp_cypto.c and ntp_protocol.c code unless you really
> understand what you are doing. We certainly will not be supporting any
> crypto which is not publically and freely and available and unencumbered
> by any patents, or other claims. In addition your crypto extension would
> have to pass a security review.
>
> Danny
>
> --
> This message has been scanned for viruses and
> dangerous content by MailScanner, and is
> believed to be clean.


Danny,
          No offense but please calm down a little.

I had talked to Dave Hart privately in the past during the week it
took me to compile ntp under Windoze with the Visual C++ Express
Edition and all the stupid hoops I had to jump through with no
"message compiler", etc..  His assumptions about Windows tools are
probably a reasonable extrapolation from previous dealings with me.
(Not that I am into Windows, I have 30 computers at home, including a
two rack 24 CPU / 37.5Gb RAM Onyx2 and some big boxes and only one
computer in the house runs Windows and I'd rather slash wrists than
write Windows code but I have to port stuff to Winblows from time to
time, its not commercially viable to me otherwise).

What I have done is really learn something about ntp_proto.c and
ntp_crypto.c
I made some really really faulty assumptions about how the existing
code worked.
Now, I as you say "really understand what you are doing."... (I am
doing)
I wish I had understood David Mills book cover to cover months before
starting this.

I probably wouldn't expect my code extension to be officially
supported. I don't think my code will ever be published or my stuff
packed into extension fields packets will see anything but government
or private networks. In fact I might never know what network my code
runs on or if the packets ever see a public network.

My code (and project) works, I learnt a lot (a real lot believe me)
and I have to start on clean up, an installer and documentation, maybe
tomorrow.

Case closed.

Thanks anyway.



Regan




More information about the questions mailing list