From owner-freebsd-arch@freebsd.org Thu Sep 17 19:23:27 2020 Return-Path: Delivered-To: freebsd-arch@mailman.nyi.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2610:1c1:1:606c::19:1]) by mailman.nyi.freebsd.org (Postfix) with ESMTP id 4C9F03EA7D4 for ; Thu, 17 Sep 2020 19:23:27 +0000 (UTC) (envelope-from jmg@gold.funkthat.com) Received: from gold.funkthat.com (gate2.funkthat.com [208.87.223.18]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client CN "gate2.funkthat.com", Issuer "Let's Encrypt Authority X3" (verified OK)) by mx1.freebsd.org (Postfix) with ESMTPS id 4Bsn1p1Nfdz4Sx7; Thu, 17 Sep 2020 19:23:25 +0000 (UTC) (envelope-from jmg@gold.funkthat.com) Received: from gold.funkthat.com (localhost [127.0.0.1]) by gold.funkthat.com (8.15.2/8.15.2) with ESMTPS id 08HJNIva026304 (version=TLSv1.2 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256 verify=NO); Thu, 17 Sep 2020 12:23:18 -0700 (PDT) (envelope-from jmg@gold.funkthat.com) Received: (from jmg@localhost) by gold.funkthat.com (8.15.2/8.15.2/Submit) id 08HJNHk5026303; Thu, 17 Sep 2020 12:23:17 -0700 (PDT) (envelope-from jmg) Date: Thu, 17 Sep 2020 12:23:17 -0700 From: John-Mark Gurney To: "Rodney W. Grimes" Cc: Brooks Davis , freebsd-arch@freebsd.org Subject: Re: Appropriate path for mount/fsck sub-programs? Message-ID: <20200917192317.GW4213@funkthat.com> Mail-Followup-To: "Rodney W. Grimes" , Brooks Davis , freebsd-arch@freebsd.org References: <20200916230922.GT4213@funkthat.com> <202009171805.08HI5W6n014921@gndrsh.dnsmgr.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <202009171805.08HI5W6n014921@gndrsh.dnsmgr.net> X-Operating-System: FreeBSD 11.3-STABLE amd64 X-PGP-Fingerprint: D87A 235F FB71 1F3F 55B7 ED9B D5FF 5A51 C0AC 3D65 X-Files: The truth is out there X-URL: https://www.funkthat.com/ X-Resume: https://www.funkthat.com/~jmg/resume.html X-TipJar: bitcoin:13Qmb6AeTgQecazTWph4XasEsP7nGRbAPE X-to-the-FBI-CIA-and-NSA: HI! HOW YA DOIN? can i haz chizburger? User-Agent: Mutt/1.6.1 (2016-04-27) X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.4.3 (gold.funkthat.com [127.0.0.1]); Thu, 17 Sep 2020 12:23:18 -0700 (PDT) X-Rspamd-Queue-Id: 4Bsn1p1Nfdz4Sx7 X-Spamd-Bar: + Authentication-Results: mx1.freebsd.org; dkim=none; dmarc=none; spf=none (mx1.freebsd.org: domain of jmg@gold.funkthat.com has no SPF policy when checking 208.87.223.18) smtp.mailfrom=jmg@gold.funkthat.com X-Spamd-Result: default: False [1.82 / 15.00]; RCVD_TLS_ALL(0.00)[]; ARC_NA(0.00)[]; RCVD_COUNT_TWO(0.00)[2]; FREEFALL_USER(0.00)[jmg]; FROM_HAS_DN(0.00)[]; RCPT_COUNT_THREE(0.00)[3]; TO_DN_SOME(0.00)[]; MID_RHS_MATCH_FROM(0.00)[]; MIME_GOOD(-0.10)[text/plain]; DMARC_NA(0.00)[funkthat.com]; AUTH_NA(1.00)[]; NEURAL_SPAM_SHORT(0.11)[0.115]; NEURAL_HAM_LONG(-0.38)[-0.379]; TO_MATCH_ENVRCPT_SOME(0.00)[]; NEURAL_HAM_MEDIUM(-0.12)[-0.116]; R_SPF_NA(0.00)[no SPF record]; FORGED_SENDER(0.30)[jmg@funkthat.com,jmg@gold.funkthat.com]; SUBJECT_ENDS_QUESTION(1.00)[]; R_DKIM_NA(0.00)[]; ASN(0.00)[asn:32354, ipnet:208.87.216.0/21, country:US]; MIME_TRACE(0.00)[0:+]; FROM_NEQ_ENVFROM(0.00)[jmg@funkthat.com,jmg@gold.funkthat.com]; MAILMAN_DEST(0.00)[freebsd-arch] X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.33 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 17 Sep 2020 19:23:27 -0000 Rodney W. Grimes wrote this message on Thu, Sep 17, 2020 at 11:05 -0700: > -- Start of PGP signed section. > > Brooks Davis wrote this message on Wed, Sep 16, 2020 at 20:42 +0000: > > > The mount and fsck commands have the ability to exec a mount_ or > > > fsck_ program to do the actual work. Currently they do this with: > > > > > > execvP(execname, _PATH_SYSPATH, argv); > > > > > > Where _PATH_SYSPATH is "/sbin:/usr/sbin" unless RESCUE is defined in > > > which case it's "/rescue:/sbin:/usr/sbin". This isn't very friendly to > > > filesystems in ports and results in things like sysutils/e2fsprogs > > > adding a link to /sbin. I think the path should be expanded to include > > > /usr/local/sbin at a minimum. If it wasn't for the /rescue hack I'd be > > > tempted to suggest just switching to `execvp` and using PATH. > > > I see a few options (semi-sorted by level of change): > > > > > > 0) Do nothing > > > 1a) Add /usr/local/sbin to _PATH_SYSPATH (it's only used in mount and > > > fsck). > > I would like to note that /usr/local is a choice made by a site > and not an absolute global constant. PREFIX can usually used > to over ride this. > > The assumption of /usr/local was once completly removed from > FreeBSD and the ports system. It has slowly crept back in > and many places now contain hard coded, not easy to override > constants. > > This is yet another proposal to hard code this non-constant. And where is this non-constant stored on the system such that it would be possible to read and add this as needed? > > > 1b) The same, but a new variable. > > > 2) Switch to _PATH_DEFPATH (that also add .../bin directories). > > > 3) Use PATH, optionally prepending /rescue > > > 4) Use exevp and teach crunched_main.c to prepend /rescue to PATH. > > > > > > I'd like opinions on what people think makes sense here. I think (0) > > > aligns poorly with current project directions, but how far to go is a > > > question to answer before implementing something. > > > > Since it's only used by fsck and mount, I'm firmly in the 1a camp.. > > > > This means that mount_* programs can be located in /usr/local/sbin as > > well, which seems to complete this change... > > > > 2 would make fsck_* programs "visible" when ports don't do the correct > > thing, which 1b helps prevent. > > > > 3 and 4 are dangerous to me, as it not everyone keeps . out of their > > path, and might be tricked into running a bad binary.. -- John-Mark Gurney Voice: +1 415 225 5579 "All that I will do, has been done, All that I have, has not."