From owner-freebsd-questions Mon Jul 8 09:13:32 1996 Return-Path: owner-questions Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id JAA15321 for questions-outgoing; Mon, 8 Jul 1996 09:13:32 -0700 (PDT) Received: from who.cdrom.com (who.cdrom.com [204.216.27.3]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id JAA15314 for ; Mon, 8 Jul 1996 09:13:29 -0700 (PDT) Received: from hermes.sees.bangor.ac.uk (hermes.sees.bangor.ac.uk [147.143.102.8]) by who.cdrom.com (8.6.12/8.6.11) with SMTP id JAA04469 for ; Mon, 8 Jul 1996 09:13:27 -0700 From: Mr D Whitehead Message-Id: <18131.9607081611@hermes.sees.bangor.ac.uk> Received: from adam.sees (adam.sees.bangor.ac.uk) by hermes.sees.bangor.ac.uk; Mon, 8 Jul 96 17:11:57 BST Subject: mountd a question and a suggestion To: questions@Freebsd.org Date: Mon, 8 Jul 1996 17:11:56 +0100 (BST) X-Mailer: ELM [version 2.4 PL23] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-questions@Freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi, I've just been configuring and testing a new machine and I've had some time consuming problems with mountd and the exports file. Firstly let me state this _is_not_a_complaint_ they are both working as documented. The problems were caused by our fairly complex export restrictions which are based on a set of (Sun based) NIS netgroups. Over time two invalid host name entries had been acquired (one typeo and one failure to remove an obsolete host name) this caused our Suns no problems, presumably the duff host names were silently ignored. The situation with the FreeBSD machine was completely different, it seems that any line in the exports file that has any kind of problem prevents mountd from exporting the intended filesystem. The problems were logged, /var/log/messages had a message stating that xxxxx was a bad line and another message stating that gethostbyname had failed. Under the circumstances while better than nothing neither message was particularly helpfull. It would have been much better if I had been told which host gethostbyname was having trouble with. Question: I was using 2.1.0-R has the code been changed/improved in later versions. Suggestion(s): 1) The gethostbyname error message should indicate what was being looked up. 2) If a gethostbyname failure occures the code should ignore this particular host (while complaining in a very loud voice) and continue to process the list so that the other hosts will have the filesystem available. I'm makeing this suggestion because even well administered sites suffer from bad typing from time to time and the loss of service that this could cause the next time the machine is rebooted could be very big. If the particular case outlined above had occured on an in service machine a large group of users would have lost access to their home directories for the 6 working hours it took me to fully revalidate the netgroup file. -- Dave Whitehead ------------------------------------------------------------------------ EMAIL:- | TELEPHONE (work):- (work) davew@sees.bangor.ac.uk | +44 1248 382703 (Direct line) (home) 100023.1076@compuserve.com | +44 1248 351151 ext 2703 ------------------------------------------------------------------------ SNAIL MAIL:- Dave Whitehead School of Electronic Engineering & Computer Systems, University College of North Wales, Dean Street, Bangor LL57 1UT ------------------------------------------------------------------------