home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.os.os2.networking
- Path: sparky!uunet!cis.ohio-state.edu!zaphod.mps.ohio-state.edu!rpi!assela
- From: assela@marcus.its.rpi.edu (A. Andre Asselin)
- Subject: Re: Killing (IBM) NFS client after last UMOUNT: dangerous?
- Message-ID: <+tk1r5+@rpi.edu>
- Nntp-Posting-Host: marcus.its.rpi.edu
- References: <341@zam103.zam.kfa-juelich.de> <1992Nov04.191754.18135@ans.net> <380@zam103.zam.kfa-juelich.de>
- Date: Sat, 7 Nov 1992 07:24:39 GMT
- Lines: 48
-
- kiehl@ibt013.ibt.kfa-juelich.de (Horst Kiehl) writes:
-
- >In article <1992Nov04.191754.18135@ans.net>,
- >db3l@ans.net (David Bolen) writes:
-
- >>In article <341@zam103.zam.kfa-juelich.de> kiehl@ibt013.ibt.kfa-juelich.de (Horst Kiehl) writes:
- >>
- >>>Is it safe to kill NFSC.EXE and the NFSBIOD.EXEs after the last UMOUNT?
- >>>(If not, how can I simulate a Ctrl+C to the NFS Control Program window?)
- >>
- >>If you are using something like DosKillProcess to shut down the
- >>processes, you won't exactly simulate a Ctrl-C (which NFSCTL uses to
- >>shut down the BIODs itself and verify that all mounts have been done),
- >>but you will still force NFSCTL to run it's exit handler, which takes
- >>care of detaching from the IFS.
-
- >I am using KillEm (from ftp-os2), which does use DosKillProcess.
- >The NFSCTL.CMD then reports the resulting RC=99 from NFSC.EXE
- >(as unknown) and exits.
-
- >Additionally, the doc from IBM says that it's safe if one closes the
- >window of the NFS Control Program when all mounted drives have been
- >unmounted beforehand.
-
- >>The gorp about checking mounts and stuff is useful, but mostly a nicety.
-
- >Well, I couldn't find what a gorp is, :-) but my aim was to run the NFS
- >client only while it is needed (because of it's overhead), i. e. only
- >"during active mounts", and that's what I reached. :-) With QMOUNT,
- >RXQUEUE, Procs & KillEm, and of course REXX, it was quite easy. :-)
-
- >> (I'm not sure if it clears all previous mounts the
- >>next time it starts up or not).
-
- >I don't think so because I remember to once have seen a message from
- >NFSC.EXE (or maybe NFSCTL.CMD) advising to run NFSCLEAN after next boot.
-
- Horst,
- The procedure you outlined is safe. As long as there are no outstanding
- mounts, it is safe to kill the NFS client. Interestingly, though, you
- mentioned that you're killing it to free the resources it uses. NFS client
- actually only uses memory when you aren't using an NFS mounted drive (so
- the only effect is a slightly bigger SWAPPER.DAT). It consumes no CPU
- time.
- You are correct on your second point, too; NFSCLEAN is used to remove any
- NFS drives that weren't properly detached for one reason or another.
- NFSCTL will not do this on its own when it's started.
- - Andre Asselin (assela@rpi.edu)
-