From: andrew@cucumber.demon.co.uk   
      
   In article <8118c585-8773-4f9f-8b7e-83ada9a1fdeb@googlegroups.com>,   
   	herrmannsfeldt@gmail.com writes:   
   > On Thursday, April 13, 2017 at 5:24:34 PM UTC-7, robert...@yahoo.com wrote:   
   >   
   > (snip, I wrote)   
   >> >What is supposed to happen if a server wants to bind() to a port   
   >> >which happens to already be in use by a TCP session?   
   >   
   >> >Since the TCP session is identified by the quad source address,   
   >> >source port, destination address, destination port, that should not   
   >> >limit the ability to bind() to that port.  But is that what   
   >> >really happens on real systems?   
   >   
   >> On most systems, you will only be able to successfully bind to a port   
   >> not otherwise being used.  There may be ways to override a bit   
   >> (SO_REUSEADDRESS/PORT), but those often don't do what you want.   
   >   
   > Interesting.  I was suspecting that it wouldn't cause a conflict,   
   > but didn't know.  It turns out that the question came from someone   
   > misunderstanding how their program worked.   
   >   
   >> In general ephemeral ports should be coming out of a different range   
   >> than statically allocated ports, and so should not conflict, OTOH,   
   >> OS's have not been consistent which ranges get assigned to what - old   
   >> Windows, for example, used to use 1025-5000 for ephemeral port, but   
   >> now uses 49152+, as do most *nix).   
   >   
   > Presumably that works unless one host wants many thousands of connections   
   > to one port on a server. Rare, but not impossible.   
      
   It comes up when trying to simulate a large connection load from   
   the Internet by using just one test system. I've seen several people   
   fall over this over the years.   
      
   It was even more of an issue with a r-protocols (rlogin, rsh, rcp)   
   where the client's local port number had to be less than 1024, but   
   those have gone away nowadays due to poor security.   
      
   --   
   Andrew Gabriel   
   [email address is not usable -- followup in the newsgroup]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)