Temporary dns on a Partitioned Network

[Martin McCormick]

	My thanks to Kevin and others on the list.  Here's what
happened for those of you who are interested.

	I finally figured out the problem or problems.  I
transplanted the named.conf file from another system and failed
to take in to account that the log file I designated was in what
was a nonexistent directory on the new system.  Fatigue can kind
of cover up little details like that.  Also, it didn't help much
not to have killed all of the instances of named that were
spawned by the previous attempt to run named.  The lack of a
working log hid that complaint and just made things silently
stick forever, or sometimes appear to work for no good reason.

	It is my understanding that most daemons start life as a
totally different process owned by whoever or whatever called it
and, after loading themselves, spawn the daemon or daemons.
Usually, you don't see it happen because it is pretty fast.  You
just start the process and see the daemon running and miss the
caller's totally different proc which goes away as soon as it has
done its job.  By trying to open a log file in /home instead of
/export/home, I botched things to the point where the caller
couldn't finish its work.

	Many thanks to those who suggested causes for the
problems.  those situations also existed so I still certainly
appreciate all the help.

Martin McCormick 405 744-7572   Stillwater, OK
OSU Center for Computing and Information services Data Communications Group

Kevin Darcy writes:
>
>Well, it's not really "hung up" if it can still resolve queries. And
>either it's in daemon mode or it isn't; there isn't any such thing as
>being "partially" in daemon mode. Also, the order in which you arrange