From owner-freebsd-questions@freebsd.org Thu Sep 3 07:29:04 2015 Return-Path: Delivered-To: freebsd-questions@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 96CE99C9768 for ; Thu, 3 Sep 2015 07:29:04 +0000 (UTC) (envelope-from raimund.sacherer@logitravel.com) Received: from formentor.toolfactory.net (pina.toolfactory.net [213.97.158.39]) by mx1.freebsd.org (Postfix) with ESMTP id B6C0D26E; Thu, 3 Sep 2015 07:29:02 +0000 (UTC) (envelope-from raimund.sacherer@logitravel.com) Received: from localhost (localhost.localdomain [127.0.0.1]) by formentor.toolfactory.net (Postfix) with ESMTP id 2AB3317833D; Thu, 3 Sep 2015 09:29:00 +0200 (CEST) Received: from formentor.toolfactory.net ([127.0.0.1]) by localhost (formentor.toolfactory.net [127.0.0.1]) (amavisd-new, port 10032) with ESMTP id DHC-t9-xcVWf; Thu, 3 Sep 2015 09:28:58 +0200 (CEST) Received: from localhost (localhost.localdomain [127.0.0.1]) by formentor.toolfactory.net (Postfix) with ESMTP id 47650178349; Thu, 3 Sep 2015 09:28:58 +0200 (CEST) X-Virus-Scanned: amavisd-new at logpmzimmta01v.toolfactory.net Received: from formentor.toolfactory.net ([127.0.0.1]) by localhost (formentor.toolfactory.net [127.0.0.1]) (amavisd-new, port 10026) with ESMTP id TJRS8HDZrJMR; Thu, 3 Sep 2015 09:28:58 +0200 (CEST) Received: from xorrigo.toolfactory.net (xorrigo.toolfactory.net [192.168.2.210]) by formentor.toolfactory.net (Postfix) with ESMTP id 2F2FC17833D; Thu, 3 Sep 2015 09:28:58 +0200 (CEST) Date: Thu, 3 Sep 2015 09:28:57 +0200 (CEST) From: Raimund Sacherer Reply-To: Raimund Sacherer To: Matthew Seaman Cc: freebsd-questions@freebsd.org Message-ID: <605077410.12419182.1441265337702.JavaMail.zimbra@logitravel.com> In-Reply-To: <55DEF1ED.4040701@FreeBSD.org> References: <938335435.8523670.1440671907486.JavaMail.zimbra@logitravel.com> <55DEF1ED.4040701@FreeBSD.org> Subject: Re: how to assure that a certain device (e.g. /dev/pass1) always is /dev/pass1 on subsequent reboots? MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [192.168.2.213] X-Mailer: Zimbra 8.0.8_GA_6184 (ZimbraWebClient - FF40 (Mac)/8.0.8_GA_6184) Thread-Topic: how to assure that a certain device (e.g. /dev/pass1) always is /dev/pass1 on subsequent reboots? Thread-Index: rIpm+4e+SoNKhAOtu0fPQwLH0Pvt3w== X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.20 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 03 Sep 2015 07:29:04 -0000 Hello, > You probably don't need to worry about this: for permanently wired > peripherals, the device names will generally be stable across reboots. > For something like an autochanger -- given they're always plugged into > the system and you can assume they're powered on -- then the devices > will be numbered in the order the kernel probes for them at boot. Just for your Info, I have one LTO5 Drive and one LTO6 Autochanger, both connected externally via pcix LSI SAS controllers. After Installation of the 2nd card (for the LTO6 Autochanger), upon reboot my devices where: pass16, sa0, LTO5 drive pass17, sa1, LTO6 drive pass18, ch0, Autochanger Yesterday I had to reboot the server and after reboot I got: pass13, sa0, LTO6 drive pass14, ch0, Autochanger pass18, sa1, LTO5 drive No hardware was touched since the installation, so there is definitly some sort of boot reordering going on. Maybe it is with the mps driver. I added the device hints so at least I think until I change the physical hardware configuration I will get the same device names. I tried it yesterday with another reboot and they stayed the same. It took me a while to figure out which driver runs the SAS controllers though, camcontrol devlist does not reference it, and I did not find it in pciconf either, but then I went trough the boot messages and found it very clearly stated. I just got confused because lspci in linux tells you the kernel driver the device is using, but the freebsd counterparts do not. Or maybe they don't show it the driver is compiled in. Well, still lot's of learning to do, thank you for your info, Ray