From owner-freebsd-current Thu Dec 21 12:52:09 1995 Return-Path: owner-current Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id MAA08010 for current-outgoing; Thu, 21 Dec 1995 12:52:09 -0800 (PST) Received: from skiddaw.elsevier.co.uk (skiddaw.elsevier.co.uk [193.131.222.60]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id MAA07999 for ; Thu, 21 Dec 1995 12:52:03 -0800 (PST) Received: from snowdon.elsevier.co.uk (snowdon.elsevier.co.uk [193.131.197.164]) by skiddaw.elsevier.co.uk (8.6.12/8.6.12) with ESMTP id UAA11057 for ; Thu, 21 Dec 1995 20:50:16 GMT Received: from cadair.elsevier.co.uk (actually host cadair) by snowdon with SMTP (PP); Thu, 21 Dec 1995 20:50:48 +0000 Received: (from dpr@localhost) by cadair.elsevier.co.uk (8.6.12/8.6.12) id UAA11834; Thu, 21 Dec 1995 20:50:45 GMT From: Paul Richards Message-Id: <199512212050.UAA11834@cadair.elsevier.co.uk> Subject: Re: Impressions of stability ( was Re: FreeBSD-current-stable ??? ) To: andreas@knobel.gun.de (Andreas Klemm) Date: Thu, 21 Dec 1995 20:50:44 +0000 (GMT) Cc: tom@smart.ruhr.de, nate@rocky.sri.MT.net, current@FreeBSD.org In-Reply-To: from "Andreas Klemm" at Dec 21, 95 07:28:44 pm X-Mailer: ELM [version 2.4 PL24] Content-Type: text Sender: owner-current@FreeBSD.org Precedence: bulk In reply to Andreas Klemm who said > > I'm currently in X11, fired up pine in an xterm. Loaded the > 900 messages long -current mailinglist folder. Edited this reply. > Edited a 20MB large messages file with vi in parallel. Truncated it > to 1M length. When I wanted to send the message I got /var filesystem > full message, first. Deceided to save the 1 MB messages file and to > leave vi. Checked space in /var. Then I got the free(): message... > > Since then cursor positioning in vi wasn't ok. Wrote the file to /tmp/xxx > (/tmp is a 90M large memory filesystem). And wrote it to /tmp/yyy. > When I entered pine this time, I wanted to include /tmp/xxx as my > reply. The file was empty (surprise, surprise). /tmp/yyy was ok, > puh... ;-) > > I think I'm surely the one and only who got this result ;-)) I'm not quite sure what the chain of events here was but it looks like something that's bitten me a few times and I really don't like the way this works. If you try and write a file to a nearly full filesystem and that write fails because the filesytem fills up, the file gets truncated. This is a very nasty thing to happen. I'm not sure what the actual course of events are but I'd really like to find some way of preventing it from happening. I'd be satisfied if I just ended up with part of the file but losing the whole thing can be really annoying if it's a chapter of your thesis, which is what once happened to me. -- Paul Richards. Originative Solutions Ltd. Internet: paul@netcraft.co.uk, http://www.netcraft.co.uk Phone: 0370 462071 (Mobile), +44 1225 447500 (work)