From owner-freebsd-hackers Thu Mar 16 06:53:21 1995 Return-Path: hackers-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id GAA10130 for hackers-outgoing; Thu, 16 Mar 1995 06:53:21 -0800 Received: from plains.NoDak.edu (tinguely@plains.NoDak.edu [134.129.111.64]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id GAA10124 for ; Thu, 16 Mar 1995 06:53:19 -0800 Received: by plains.NoDak.edu; Thu, 16 Mar 1995 08:53:10 -0600 Date: Thu, 16 Mar 1995 08:53:10 -0600 From: Mark Tinguely Message-Id: <199503161453.AA07760@plains.NoDak.edu> To: hackers@FreeBSD.org, phk@ref.tfs.com Subject: Re: fsck buglet Sender: hackers-owner@FreeBSD.org Precedence: bulk > If I say "fsck -n /dev/rfd0", wouldn't it make sense for fsck to return > a non-zero exit code if any trouble were found ? > > As it is it will return zero no matter what it finds... philosophically, I think only the preen option ("-p") should return a status. I can see that you want to check out floppies, but people will also run fsck on mounted filesystem. If the filesystem is mounted, then at least the "clean flag" in the superblock will be wrong giving you the nonzero return status. if you change "fsck -n" that a "problem" exists, (honest question not trying to be a smart-ass) then should you add a return code that a "problem" was solved with "fsck [-y | -p | ]" to be consistant? --mark.