Sockets that dont know when to leave...

[Andy S Poling]

Hmm.  We just upgraded a uVax II to Ultrix 4.1 rev 52 (which included a
mandatory update) and are now having a rather strange problem with UDP
sockets...

When certain daemons (UREP's SNA-over-UDP daemon for instance) are killed,
the UDP sockets that they opened don't die with them.  This is a BIG problem
because the UDP socket these daemons use must be bound to a certain local
address (port).  So once a daemon has run and has either exited or has been
killed that particular port is no longer available for binding to a socket
(we get EADDRINUSE - errno 48).  When one uses "netstat -a" to see what
sockets are in use there will indeed be a UDP socket open with that address
assigned.  Furthermore netstat usually shows that there is data in the recv
queue for that socket.

To me this does not seem correct - and it is not consistent: it does not
always happen.  It was not a problem with our previous MUCH older version of
Ultrix.  Am I wrong?  What can be done to eradicate a socket which belongs
to noone (short of rebooting...)?

-Andy

PS: don't use the BITNET mail address below.  It won't reach me until we get
this problem fixed...

--
Andy Poling                              Internet: andy at jhunix.hcf.jhu.edu
UNIX Systems Programmer                  Bitnet: ANDY at JHUNIX
Homewood Academic Computing              Voice: (301)338-8096    
Johns Hopkins University                 UUCP: uunet!mimsy!aplcen!jhunix!andy