[15028] in Athena Bugs
Re: sgi [8.0K]: rpc.lockd
daemon@ATHENA.MIT.EDU (John Hawkinson)
Fri Mar 21 09:42:44 1997
Date: Fri, 21 Mar 1997 09:42:41 -0500 (EST)
To: Greg Hudson <ghudson@MIT.EDU>
Cc: Jonathon Weiss <jweiss@MIT.EDU>, bugs@MIT.EDU
In-Reply-To: "[15026] in Athena Bugs"
From: John Hawkinson <jhawk@MIT.EDU>
| Finding a solution to the problem that does the minimum amount of
| violence to /etc/rc2.d and doesn't break anything will take some
| analysis, and I don't guarantee that it will happen in time for the
| 8.1 release if the task is left to me.
Perhaps this was completely clear and doesn't need to be restated,
but it appears that is just as much a Solaris bug as an Athena
bug, and thus should be reported to Sun.
Solaris should:
Start inetd prior to rpcbind
Make it very easy for custom daemons to start
before rpcbind (be they inetd-like daemons such
as the athena inetd, or just more conventional
daemons which use a non-reserved port).
Further, as a bit of irony, rpcbind is failing to respect
the conventions designated by RFC1700, viz.:
| The Well Known Ports are those from 0 through 1023.
| The Registered Ports are those from 1024 through 49151
| The Dynamic and/or Private Ports are those from 49152 through 65535
Is there someone involved in forwarding bug reports from bugs to Sun?
I always got the impression the answer was No, so I suspect we'll lose
here. Perhaps one of the benevolent monarchs^H^H^H^H^H^H^H^Hpeople-
with-Sun-bits will pop up at some point.
--jhawk
(who feels like he's losing the battle if we move random athena services
to the Dynamic range only to move the rpcbind stuff there too, alas alack,
sometimes this just sucks)
