From owner-freebsd-questions Wed Oct 2 13:37:33 2002 Delivered-To: freebsd-questions@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id DAFDA37B401 for ; Wed, 2 Oct 2002 13:37:29 -0700 (PDT) Received: from web40511.mail.yahoo.com (web40511.mail.yahoo.com [66.218.78.128]) by mx1.FreeBSD.org (Postfix) with SMTP id 933E443E65 for ; Wed, 2 Oct 2002 13:37:29 -0700 (PDT) (envelope-from dannypansters@yahoo.com) Message-ID: <20021002203729.67307.qmail@web40511.mail.yahoo.com> Received: from [217.121.1.115] by web40511.mail.yahoo.com via HTTP; Wed, 02 Oct 2002 13:37:29 PDT Date: Wed, 2 Oct 2002 13:37:29 -0700 (PDT) From: Danny Pansters Subject: Quite OT, problems with AOpen DX34 Plus motherboard and disks being trashed To: freebsd-questions@freebsd.org MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG I thought, since I run FreeBSD on this particular server box, I might try posting here also. Sorry about length of post but it still doesn't have all info I could deliver as it is. As posted on motherboard.org forum: ------ Hi. I could really use some advice here. The mobo has two PIII-800s, onboard Adaptec scsi controller, Via chipset. Was using 2 identical Seagate Barracuda 18 GB disks with software RAID1. OS: FreeBSD 4.6. When referring to "first" disk: this is the one I boot the OS off. Now to the horror story... After approx. 1.5 years of running virtually 24/7 my first SCSI disk got read and write errors, and upon rescanning it would have disappeared from the chain. Upon rebooting the same would re-occur after non-equal but overall increasingly shorter periods. In the end: hours. Occasionally the drive wouldn't be seen by SCSI BIOS at all. The other one (at the same chain) would. First thought: drive is dying. However when using Seagate's diagnostic boot disk the drive would either not show up, or it would and if it did it passed all tests. Second thought: try other disk anyway. Because I want to keep working with a mirrored configuration I got two new identical disks, ATA this time (WD, 20GB) with a new and short ATA100 cable. Dumped my data and OS from the second SCSI drive and recreated configuration. Incidentally, once I had everything running, I could mount the other (supposedly faulty) SCSI disk and read/write just fine. Within days though, the second ATA disk started to show the same problems as my former first SCSI disk did earlier. Again upon rebooting it would work again but not very long. Incidentally clicking noises. Not good! Third thought: bad power supply. From what I've read, opinions on this vary, but I came across enough posts in mailing lists archives (FreeBSD's and some others) indicating that a bad supply can do a lot of damage. So I got me top knotch Antec power supply and tossed the old one out. Connected system fan to it and things seemed fine. At least the box was a lot less noisy (the power supply itself has not one but two fans also of itself effectively pushing the warm air out of the server enclosure plus it can regulate system fan speed when needed). However, within a day or two my system logs again showed read errors. And more ticking and clicking. Outch. Fourth thought: check temperature, voltages, airflow, etc. Actually, I ahve been looking at operational variables like these all along. With the new power supply, mobo temp would go to some 35 degrees celcius, with the old one up to 39. Not too much. As for voltages, well, they looked normal, except that the -12V would always be close to -17. I am not sure if this is wrong, AFAIK it has always been like that. It certainly didn't change when I replaced the power supply. Also, none of the disks ever appeared to become very warm. Fifth thought, perhaps getting more far fetched: for some reason, the offending disk always seemed to be the one physically located at the bottom of the 5.25" cage. I probably need to explain what this thing looks like: The cage can be taken out of the server enclosure wholly, it holds the CDROM, and one can screw up to five 3.5" devices around it. The most logical place to start is using the positions under the CDROM drive and to make them fit one has to use the metal (I think copper) "strips" on both sides screwed into the harddisks and into the cage. These came with the machine and were indeed used to hold the harddisks. Other places to put additional disks are on top and on both sides of the cage forming kind of a cube altogether. Yes, I have made VERY sure that the disk at the bottom is not touching another disk, neither does it touch the underside of the cage or the enclosure. The things is, a ticking ATA disk is often caused by too low voltage which wears it out quickly, as far as I've understood. Could I be having some problem with grounding? And where then, how could this have suddenly started? Otherwise, well, I can't think of anything else than to replace the mobo. Can a damaged mobo lead to problems as described above? It would have to happen somewhere "upstream" from the ATA/SCSI controllers then. In short, I'm really stumped on this! Last time I tried to power up the damn thing all it did was click, it didn't even get to post, let alone boot :( I'm very reluctant to keep on trying and testing because in the meantime chances are my disks will deteriorate further (I suspect the ATA drive is already toast). Any ideas, recommendations, experiences that might help me?? Thanks a ton for even reading this long post. __________________________________________________ Do you Yahoo!? New DSL Internet Access from SBC & Yahoo! http://sbc.yahoo.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message