From owner-freebsd-hackers@freebsd.org Wed Mar 8 14:01:29 2017 Return-Path: Delivered-To: freebsd-hackers@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 11BD7D033D7 for ; Wed, 8 Mar 2017 14:01:29 +0000 (UTC) (envelope-from bob@immure.com) Received: from maul.immure.com (108-84-10-9.lightspeed.austtx.sbcglobal.net [108.84.10.9]) by mx1.freebsd.org (Postfix) with ESMTP id E4CD31DD3 for ; Wed, 8 Mar 2017 14:01:28 +0000 (UTC) (envelope-from bob@immure.com) Received: from rancor.immure.com ([10.1.132.9]) by maul.immure.com with esmtp (Exim 4.88 (FreeBSD)) (envelope-from ) id 1clc9e-000JCl-9R; Wed, 08 Mar 2017 08:01:27 -0600 Received: from rancor.immure.com (localhost [127.0.0.1]) by rancor.immure.com (8.15.2/8.15.2) with ESMTP id v28E1P6I017243; Wed, 8 Mar 2017 08:01:25 -0600 (CST) (envelope-from bob@rancor.immure.com) Received: (from bob@localhost) by rancor.immure.com (8.15.2/8.14.9/Submit) id v28E1PxJ017242; Wed, 8 Mar 2017 08:01:25 -0600 (CST) (envelope-from bob) Date: Wed, 8 Mar 2017 08:01:25 -0600 From: Bob Willcox To: "Rodney W. Grimes" Cc: hackers list Subject: Re: Help with silent reboot of 10.3-stable system Message-ID: <20170308140125.GF22199@rancor.immure.com> Reply-To: Bob Willcox References: <20170308015630.GE22199@rancor.immure.com> <201703080541.v285ftuW098706@pdx.rh.CN85.dnsmgr.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <201703080541.v285ftuW098706@pdx.rh.CN85.dnsmgr.net> User-Agent: Mutt/1.7.1 (2016-10-04) X-BeenThere: freebsd-hackers@freebsd.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Technical Discussions relating to FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 08 Mar 2017 14:01:29 -0000 On Tue, Mar 07, 2017 at 09:41:55PM -0800, Rodney W. Grimes wrote: > > Over the past month or so my network fileserver system (NFS support for my > > entire, small, network) has begun silently rebooting itself. Here is the uname > > -a output: > > > > FreeBSD vader.immure.com 10.3-STABLE FreeBSD 10.3-STABLE #15 r313997: Mon Feb 20 14:40:00 CST 2017 bob@vader.immure.com:/usr/obj/usr/src/sys/GENERIC amd64 > > > > At first I suspected that it might be the power supply as it was a couple of > > years old so I replaced that. Unfortunately, it has begun doing it again (had > > a couple of weeks respite) so now my suspicions seem to have been incorrect. > > > > I was hoping that someone might be able to give me some clues on what I can do > > to reveal the problem. Are there any general debug settings for the kernel (or > > elsewhere) that would maybe give an indication of why it is being rebooted > > (assuming it's a software problem)? > > > > Thanks for any suggestions you may have! > > Given that you have already suspected hardware I'll continue down that > road and leave the software road for others to persue. > > Was it rebooting more often than every couple of weeks? It sounds > as if a power supply swap fixed the problem for a short period, > but it has come back. If that is true my suspecion would be > bad primary side filter caps in the cpu vrm on the motherboard. > > Your replacment powersupply has nice new filter caps, if you > did put in a new power supply, if you put in a used one, go > get a brand now one. PC power supplies are junk when it comes > to there output filter stages, and I dont care how expenive > of a supply you buy. No one engineers a life of more than > 3 years into them anymore. Anyway, changing this cleaned > up the primary side of the vrm for a while, but since those > capacitors are degrading this let them get even worse as > when caps start leaking they make heat and the hotter they > get the more they leak and it spirals into a cook off that > usually ends in the cap leaking blank gunk out the top, > or in some cases out the bottom. > > Please look very carefully at the MB CPU filter caps, google can help you > if your now a hardware type to find what your looking for. > Google: motherboard bad caps > > How old is the is motherboard? Anything more than 2 years old can easily > have degraded caps. Unless it is using solid polymer types, then I give > them 3 or 4 years. Again nothing is engineered to last much beyond > warranty. Any life beyond warranty is not by design, but simply the > accidental nature of things often work better than speced. > > > Bob > > -- > > Bob Willcox | If a program is useful, it will be changed. > > bob@immure.com | > > Austin, TX | > > _______________________________________________ > > freebsd-hackers@freebsd.org mailing list > > https://lists.freebsd.org/mailman/listinfo/freebsd-hackers > > To unsubscribe, send any mail to "freebsd-hackers-unsubscribe@freebsd.org" > > -- > Rod Grimes rgrimes@freebsd.org Thanks for the reply Rod. My fear is that you may well be right. This system is about two years old now and it had been running flawlessly right up until last month with if first started doing what appears to be a temporary shutdown of power. The power supply that I installed recently is a new one (over the years I have become quite distrustful of power supplies in general). Unfortunately, replacing the motherboard is going to be painful as this is a Silverstone 8 drive NAS case (DS380B), and although a great case, it leaves very little room to work inside (and it's full). Also, being my networks main file server I depend on it almost totally so I'm reluctant to start tearing it apart. I may bite the bullet and build a complete replacement system while this one is still (mostly) working. Quite expensive, but seems it will be the safest path. Bob -- Bob Willcox | If a program is useful, it will be changed. bob@immure.com | Austin, TX |