From owner-freebsd-current Sun Nov 1 01:34:52 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id BAA03804 for freebsd-current-outgoing; Sun, 1 Nov 1998 01:34:52 -0800 (PST) (envelope-from owner-freebsd-current@FreeBSD.ORG) Received: from marvin.spacequest.hs (in17.fto.de [193.197.153.146]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id BAA03798 for ; Sun, 1 Nov 1998 01:34:49 -0800 (PST) (envelope-from hschaefer@fto.de) Received: from daneel.spacequest.hs (daneel.spacequest.hs [192.168.0.98]) by marvin.spacequest.hs (8.8.8/8.8.7) with ESMTP id VAA25122; Sat, 31 Oct 1998 21:40:48 +0100 (CET) (envelope-from hschaefer@fto.de) Date: Sat, 31 Oct 1998 21:39:38 +0100 (CET) From: Heiko Schaefer X-Sender: heiko@daneel.spacequest.hs To: Heiko Schaefer cc: freebsd-current@FreeBSD.ORG Subject: Re: Problem with SCSI Harddisk In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Hi FreeBSD-current comunity again, as no one answered to my question (that i posted here a week ago), i figured out that either this is the wrong place to ask, or no one found this important or no one had anything to say about it. ...just in case anyone has a similar problem sometime, i wanted to write about what i tried, and how i finally found a 'workaround' to my problem today (though it feels a bit funny replying to one's own mails :)). here's the problem once again: > I am using FreeBSD 3.0-CURRENT (of today), and have a problem with one of > my scsi-harddisks. /var/log/messages says: > > Oct 24 10:00:56 daneel /kernel: (da3:ahc0:0:3:0): SCB 0xe1 - timed out > while idle, LASTPHASE == 0x1, SCSISIGI == 0x0 > Oct 24 10:00:56 daneel /kernel: SEQADDR == 0xb > Oct 24 10:00:56 daneel /kernel: SSTAT1 == 0xa > Oct 24 10:00:56 daneel /kernel: (da3:ahc0:0:3:0): Queuing a BDR SCB > Oct 24 10:00:56 daneel /kernel: (da3:ahc0:0:3:0): Bus Device Reset Message > Sent > Oct 24 10:00:56 daneel /kernel: (da3:ahc0:0:3:0): no longer in timeout, > status = 34b > Oct 24 10:00:56 daneel /kernel: ahc0: Bus Device Reset on A:3. 64 SCBs > aborted > Oct 24 10:00:56 daneel /kernel: (da3:ahc0:0:3:0): tagged openings now 64 > > > this message is then repeated for zillions of times (exactly once every > 60 seconds), once it has occured once (it occurs for the first time, when > the harddisk is used a bit more heavily). eventually the whole system > hangs (responds to ping, but not to telnet or anything similar, although > the filesystem on the harddisk contains no part of the system, just data > of a samba-fileserver). i am now quite sure, that the problem started occuring, when i switched from 2.2-STABLE to 3.0-CURRENT (i.e. from the old scsi driver to cam ?!). the first thing i tried to get rid of the messages (and system hangs), i downgraded my system from 3.0-CURRENT (of about 25.10.98) to 3.0-RELEASE by cvsup'ing and making world and a new kernel, but that did not help. then i decided to use another scsi controller. after replacing the 2940UW (a bit older) with a 2940U (quite new), the problem seems to have disappeard. the problem occured when using this one: > ahc0: rev 0x00 int a irq 10 on pci0.12.0 > ahc0: aic7880 Wide Channel A, SCSI Id=7, 16/255 SCBs (but funnily only with one of my four harddisks - a cheap 5 1/4" seagate 9GB harddisk: da3: ... somehow the 2940UW didn't 'like' this harddisk ?! :> ). the controller that i now use is: ahc0: rev 0x01 int a irq 10 on pci0.12.0 ahc0: aic7860 Single Channel A, SCSI Id=7, 3/255 SCBs maybe this is a known problem ?! if not, is it likely to be a hardware defect of some kind, or is it a bug in freebsd ? am i supposed (would it help) to report this to some developer (or are all developers reading this list ?) have a nice day, Heiko To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message