From owner-freebsd-scsi@FreeBSD.ORG Fri Aug 11 15:43:54 2006 Return-Path: X-Original-To: freebsd-scsi@freebsd.org Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 00A0A16A4DD for ; Fri, 11 Aug 2006 15:43:54 +0000 (UTC) (envelope-from geoffb@chuggalug.clues.com) Received: from chuggalug.clues.com (chuggalug2.demon.co.uk [83.104.169.191]) by mx1.FreeBSD.org (Postfix) with ESMTP id 5016043D49 for ; Fri, 11 Aug 2006 15:43:52 +0000 (GMT) (envelope-from geoffb@chuggalug.clues.com) Received: from chuggalug.clues.com (localhost [127.0.0.1]) by chuggalug.clues.com (8.12.10/8.12.10) with ESMTP id k7BFhmKP084148; Fri, 11 Aug 2006 15:43:48 GMT (envelope-from geoffb@chuggalug.clues.com) Received: (from geoffb@localhost) by chuggalug.clues.com (8.12.10/8.12.10/Submit) id k7BFhmkw084147; Fri, 11 Aug 2006 15:43:48 GMT (envelope-from geoffb) Date: Fri, 11 Aug 2006 15:43:48 +0000 From: Geoff Buckingham To: Eric Anderson Message-ID: <20060811154348.GA83765@chuggalug.clues.com> References: <44DB8A9C.8090609@centtech.com> <44DC6F9F.4060405@centtech.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <44DC6F9F.4060405@centtech.com> User-Agent: Mutt/1.4.1i Cc: freebsd-scsi@freebsd.org Subject: Re: isp issues on recent -STABLE X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 11 Aug 2006 15:43:54 -0000 On Fri, Aug 11, 2006 at 06:53:03AM -0500, Eric Anderson wrote: > [..snip..] > Aug 9 23:02:10 snapshot1 kernel: isp0: command timed out for 0.2.2 > Aug 9 23:02:10 snapshot1 kernel: (da8:isp0:0:2:2): Command timed out > Aug 9 23:02:10 snapshot1 kernel: (da8:isp0:0:2:2): Retrying Command > Aug 9 23:26:58 snapshot1 kernel: (da3:isp0:0:1:0): Queue Full > Aug 9 23:26:58 snapshot1 kernel: (da3:isp0:0:1:0): tagged openings now 254 I don't what may have changed in the driver recently, but from your post you seem to be using the FC isp and potentially a "SAN" presenting arrays as luns to you rather than JBOD on a loop or switch. If you you have some kind of data mover presenting arrays, your SAN vendor may well recomend a maximum queue size per lun (often 20-30) man camcontrol, look at the tags section. Your commands may be timing out because you have managed to queue too many command (which I hope should not happen). Or..... Your queues could be filling because your commands are timing out. Which would imply something is broke :-(