From owner-freebsd-questions@freebsd.org Fri Oct 7 16:42:10 2016 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 1392CBED7E2 for ; Fri, 7 Oct 2016 16:42:10 +0000 (UTC) (envelope-from robroy@robroygregg.com) Received: from mail.robroygregg.com (173-13-147-189-sfba.hfc.comcastbusiness.net [173.13.147.189]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id D6D02BFD for ; Fri, 7 Oct 2016 16:42:09 +0000 (UTC) (envelope-from robroy@robroygregg.com) Received: from funmax (funmax.d.net [192.168.16.3]) by mail.robroygregg.com (OpenSMTPD) with ESMTP id e370d0b1; Fri, 7 Oct 2016 09:42:07 -0700 (PDT) Date: Fri, 7 Oct 2016 09:42:07 -0700 (PDT) From: Robroy Gregg X-X-Sender: robroy@funmax.d.net To: Vijay Rajah cc: freebsd-questions@freebsd.org Subject: Re: isp(4) QLE2462 initiator failure with 10.3-RELEASE In-Reply-To: <0a7120b7-54d2-5afe-5e9a-3681e74d543e@rvijay.me> Message-ID: References: <0a7120b7-54d2-5afe-5e9a-3681e74d543e@rvijay.me> User-Agent: Alpine 2.20 (BSF 67 2015-01-07) MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII; format=flowed X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 07 Oct 2016 16:42:10 -0000 On Fri, 7 Oct 2016, Vijay Rajah wrote: > On 10/06/2016 10:50 PM, Robroy Gregg wrote: >> * I've tried physically swapping the isp0 and isp1 HBA port >> connections--the symptom stuck to isp0, even when its I/Os were >> being attempted through the physical connection formerly used >> (successfully) by isp1. > > This might be an indication of an issue with the Motherboard (MB)(more > specifically issue with the PCIE slot -- Since the problem stays with the > slot). Thank you for replying Vijay! Yeah, though isp0 and isp1 are combined on a single card, in the same physical slot, I understand that they act like two separate devices at some level. > Do you have another PCIE (x4 or higher) slot in the motherboard? if so, > can you use that unsued slot to see if that solves the problem. Yes, there are one or two other empty slots to try. The trouble is, this event has happened only once so far, and we haven't been able to figure out what triggered it. So the new slot approach will be conclusive only if the problem happens again, after the slot's been changed. > OR alternatively, Do you have simllar MB's in your inventory or servers using > the same MB's? if so, can you swap the MB's and see if the issues followes > the MB? (meaning, does the issue occur in the swapped server) We do have a second computer that's identical to this one other than having less RAM, and we'll be bringing that one up soon. It'll be interesting to see if the problem happens there. I'm hoping that somebody familiar with isp(4)'s guts may be able to collect information about the card's bizarre state, and make a stab at what may have caused this. Sadly, I think I have to reboot the computer to enable hint.isp.X.debug, which will probably cause the problem to run away and hide indefinitely. I'd also love to know how many other FreeBSD users are having success with similar configurations. Do QLE2462s work flawlessly for other SAN users? It's ironic that people keep quiet both when hardware X works, and when they're not using hardware X at all. It's hard to judge how esoteric our configuration is for FreeBSD. > Hope this helps I really appreciate your reply Vijay; thank you again. Robroy Gregg Salinas, California