On 19 Nov 2008, at 01:00,  "Michael Ash" <[EMAIL PROTECTED]> wrote:


On Tue, Nov 18, 2008 at 4:49 AM, Gerriet M. Denkmann
<[EMAIL PROTECTED]> wrote:

On 18 Nov 2008, at 15:27, Stefan Werner wrote:

Aren't port numbers 0-1023 privileged under Unix and therefore only to applications running as root? If you're running your application as a regular user, it can't open a listening socket on port 123. It should still be perfectly able to send requests to port 123 on other hosts and receive
answers to those requests.

That is exactly what I am trying to do (and which worked perfectly on
Tiger): send some message on port 123 (to some ntp server) and receive an
answer.
On Leopard I send, but never get an answer.
It does not seem to be an endian issue (the running version was last tested
on a Tiger iBook ppc) - both intel and ppc produce the same 48 byte
packages.

I really don't know what to try next.

Use Wireshark or another sniffer of your choice to find out what's
actually happening on the network.

In an other programm (doing Distributed Objects between computers) I found out that int y = bind( newSocket, (struct sockaddr *)&serverAddress, sizeof(serverAddress) ); wants to have the data in "truct sockaddr_in serverAddress" in network byte order (aka big endian) (did see no mention of this in the man page for bind, nor in "in.h" or "socket.h").
Like:
        serverAddress.sin_port                  = htons(portNbr);       
        serverAddress.sin_addr.s_addr   = htonl(INADDR_ANY);
With these changes, this other programm is now working again.



You have too many variables right now.

Absolutely true.
In order to check whether it is a Tiger/Leopard issue or an endian one (I works on Ppc Tiger), I tried to install Tiger into an empty partition, but failed miserably. My dealer told me that newer hardware (in this case a white MacBook bought last week) cannot boot in Tiger.
Can this be true?
If yes, can it be circumvented?
And: is there some valid technical reason for this, or is it just some silly marketing thing?

Anyway, my current guess is: the problem is an endian issue.
If I fail, I will look for Wireshark. But: what is this? Some hardware device? Or software?

Did it get sent? Did a reply
get sent? Did your computer block the reply, or the sending, or did
the remote computer not like it, or...? Wireshark will tell you what
is actually happening so you don't have to guess, and then at that
point hopefully the solution will come more easily.

Gerriet.

_______________________________________________

Cocoa-dev mailing list (Cocoa-dev@lists.apple.com)

Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com

Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com

This email sent to [EMAIL PROTECTED]

Reply via email to