From owner-freebsd-scsi@FreeBSD.ORG Sat Feb 19 09:23:46 2005 Return-Path: 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 6C8F416A4CE for ; Sat, 19 Feb 2005 09:23:46 +0000 (GMT) Received: from svm.csie.ntu.edu.tw (svm.csie.ntu.edu.tw [140.112.90.75]) by mx1.FreeBSD.org (Postfix) with ESMTP id 92B7943D54 for ; Sat, 19 Feb 2005 09:23:45 +0000 (GMT) (envelope-from rafan@svm.csie.ntu.edu.tw) Received: from svm.csie.ntu.edu.tw (localhost [127.0.0.1]) by svm.csie.ntu.edu.tw (8.13.1/8.13.1) with ESMTP id j1J9N7PX095481; Sat, 19 Feb 2005 17:23:07 +0800 (CST) (envelope-from rafan@svm.csie.ntu.edu.tw) Received: (from rafan@localhost) by svm.csie.ntu.edu.tw (8.13.1/8.13.1/Submit) id j1J9N0kG095464; Sat, 19 Feb 2005 17:23:00 +0800 (CST) (envelope-from rafan) Date: Sat, 19 Feb 2005 17:23:00 +0800 From: Rong-En Fan To: Xin LI Message-ID: <20050219092300.GA95182@svm.csie.ntu.edu.tw> References: <20050204102558.GA2001@svm.csie.ntu.edu.tw> <1108667040.656.20.camel@spirit> <20050218071746.GA41419@svm.csie.ntu.edu.tw> <20050218202901.GA71842@svm.csie.ntu.edu.tw> <20050219061517.GA92149@svm.csie.ntu.edu.tw> <1108803317.602.13.camel@spirit> Mime-Version: 1.0 Content-Type: text/plain; charset=gb2312 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <1108803317.602.13.camel@spirit> User-Agent: Mutt/1.5.7i cc: Rong-En Fan cc: ob@e-Gitt.NET cc: scsi@freebsd.org Subject: Re: Problem with mpt(4) and Infortrend RAID X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 19 Feb 2005 09:23:46 -0000 On Sat, Feb 19, 2005 at 04:55:17PM +0800, Xin LI wrote: > Is it possible to get the first error message? The first one is usually below is the first error message... :-( > more useful. Another thing that is worthy to do is try to reduce the > allowed tag depth, through camcontrol(8): > > camcontrol tags da0 -N 32 > camcontrol tags da1 -N 32 This server is now in production mode, and is in 160MB/s since crash few hours ago (see my previous mail). Unfortunately, I saw similar message when I'm trying newfs da1s1d (1024G): The first error message is: Feb 19 16:07:10 kernel: mpt1: time out on request index = 0xfd sequence = 0x00000af5 Feb 19 16:07:10 kernel: mpt1: Status 00000001; Mask 00000001; Doorbell 24000000 Feb 19 16:07:10 kernel: request state On Chip Feb 19 16:07:10 kernel: SCSI IO Request @ 0xe6489c20 Feb 19 16:07:10 kernel: Chain Offset 0x10 Feb 19 16:07:10 kernel: MsgFlags 0x00 Feb 19 16:07:10 kernel: MsgContext 0x000000fd Feb 19 16:07:10 kernel: Bus: 0 Feb 19 16:07:10 kernel: TargetID 0 Feb 19 16:07:10 kernel: SenseBufferLength 32 Feb 19 16:07:10 kernel: LUN: 0x0 Feb 19 16:07:10 kernel: Control 0x01000000 WRITE SIMPLEQ Feb 19 16:07:10 kernel: DataLength 0x00010000 Feb 19 16:07:10 kernel: SenseBufAddr 0x7d400be0 Feb 19 16:07:10 kernel: CDB[0:10] 2a 00 45 0a fb 5f 00 00 80 00 Feb 19 16:07:10 kernel: SE32 0xe676ea30: Addr=0x35f79000 FlagsLength=0x14001000 Feb 19 16:07:10 kernel: HOST_TO_IOC Feb 19 16:07:10 kernel: SE32 0xe676ea38: Addr=0x434ba000 FlagsLength=0x94001000 Feb 19 16:07:10 kernel: HOST_TO_IOC LAST_ELEMENT Feb 19 16:07:10 kernel: CE32 0xe676ea40: Addr=0x7d400a48 NxtChnO=0x16 Flgs=0x30 Len=0x60 Feb 19 16:07:10 kernel: SE32 0xe676ea48: Addr=0x44d3b000 FlagsLength=0x14001000 Feb 19 16:07:10 kernel: HOST_TO_IOC Feb 19 16:07:10 kernel: SE32 0xe676ea50: Addr=0x4551c000 FlagsLength=0x14001000 Feb 19 16:07:10 kernel: HOST_TO_IOC Feb 19 16:07:10 kernel: SE32 0xe676ea58: Addr=0x4595d000 FlagsLength=0x14001000 Feb 19 16:07:10 kernel: HOST_TO_IOC Feb 19 16:07:10 kernel: SE32 0xe676ea60: Addr=0x4367e000 FlagsLength=0x14001000 Feb 19 16:07:10 kernel: HOST_TO_IOC Feb 19 16:07:10 kernel: SE32 0xe676ea68: Addr=0x4585f000 FlagsLength=0x14001000 Feb 19 16:07:10 kernel: HOST_TO_IOC Feb 19 16:07:10 kernel: SE32 0xe676ea70: Addr=0x36f60000 FlagsLength=0x14001000 Feb 19 16:07:10 kernel: HOST_TO_IOC Feb 19 16:07:10 kernel: SE32 0xe676ea78: Addr=0x46581000 FlagsLength=0x14001000 Feb 19 16:07:10 kernel: HOST_TO_IOC Feb 19 16:07:10 kernel: SE32 0xe676ea80: Addr=0x428a2000 FlagsLength=0x14001000 Feb 19 16:07:10 kernel: HOST_TO_IOC Feb 19 16:07:10 kernel: SE32 0xe676ea88: Addr=0x45823000 FlagsLength=0x14001000 Feb 19 16:07:10 kernel: HOST_TO_IOC Feb 19 16:07:10 kernel: SE32 0xe676ea90: Addr=0x45aa4000 FlagsLength=0x14001000 Feb 19 16:07:10 kernel: HOST_TO_IOC Feb 19 16:07:10 kernel: SE32 0xe676ea98: Addr=0x43d05000 FlagsLength=0x94001000 Feb 19 16:07:10 kernel: HOST_TO_IOC LAST_ELEMENT Feb 19 16:07:10 kernel: CE32 0xe676eaa0: Addr=0x7d400aa8 NxtChnO=0x0 Flgs=0x30 Len=0x18 Feb 19 16:07:10 kernel: SE32 0xe676eaa8: Addr=0x43e86000 FlagsLength=0x14001000 Feb 19 16:07:10 kernel: HOST_TO_IOC Feb 19 16:07:10 kernel: SE32 0xe676eab0: Addr=0x35da7000 FlagsLength=0x14001000 Feb 19 16:07:10 kernel: HOST_TO_IOC Feb 19 16:07:10 kernel: SE32 0xe676eab8: Addr=0x450c8000 FlagsLength=0xd5001000 Feb 19 16:07:10 kernel: HOST_TO_IOC LAST_ELEMENT END_OF_BUFFER END_OF_LIST If you need rest ones, I can mail you :-) I just noticed that in my RAID console, there are some parameters for host-side: Maximum Queued I/O Count - 256 LUNs per Host SCSI ID - 8 Max Number of Concurrent Host-LUN Connection - Def(4) Number of Tags Reserved for each Host-LUN Connection - Def(32) Peripheral Device Type Parameters Host Cylinder/Head/Sector Mapping Configuration I'm not that similar with SCSI, if it possible some parameters here need to tune? Regards, rafan > ÔÚ 2005-02-19ÁùµÄ 14:15 +0800£¬Rong-En FanдµÀ£º > > I was wrong, with patch, and 1 LD -> 1 channel, 320MB/s > > after 12hours, I got this: > > > > mpt0: time out on request index = 0xf6 sequence = 0x00246948 > > mpt0: Status 00000001; Mask 00000001; Doorbell 24000000 > > request state On Chip > > SCSI IO Request @ 0xe6489c20 > > Chain Offset 0x10 > > MsgFlags 0x00 > > MsgContext 0x000000f6 > > Bus: 0 > > TargetID 0 > > SenseBufferLength 32 > > LUN: 0x0 > > Control 0x01000000 WRITE SIMPLEQ > > DataLength 0x00010000 > > SenseBufAddr 0x7d610de0 > > CDB[0:10] 2a 00 6c 59 d7 50 00 00 80 00 > > SE32 0xe653dc30: Addr=0x3721e000 FlagsLength=0x14001000 > > HOST_TO_IOC > > SE32 0xe653dc38: Addr=0x7a65f000 FlagsLength=0x94001000 > > HOST_TO_IOC LAST_ELEMENT > > CE32 0xe653dc40: Addr=0x7d610c48 NxtChnO=0x16 Flgs=0x30 Len=0x60 > > SE32 0xe653dc48: Addr=0x76380000 FlagsLength=0x14001000 > > HOST_TO_IOC > > SE32 0xe653dc50: Addr=0xbae1000 FlagsLength=0x14001000 > > HOST_TO_IOC > > SE32 0xe653dc58: Addr=0x559a2000 FlagsLength=0x14001000 > > HOST_TO_IOC > > SE32 0xe653dc60: Addr=0x5343000 FlagsLength=0x14001000 > > HOST_TO_IOC > > SE32 0xe653dc68: Addr=0x282c4000 FlagsLength=0x14001000 > > HOST_TO_IOC > > SE32 0xe653dc70: Addr=0x41c5000 FlagsLength=0x14001000 > > HOST_TO_IOC > > SE32 0xe653dc78: Addr=0x231c6000 FlagsLength=0x14001000 > > HOST_TO_IOC > > SE32 0xe653dc80: Addr=0x5e9e7000 FlagsLength=0x14001000 > > HOST_TO_IOC > > SE32 0xe653dc88: Addr=0x2ffa8000 FlagsLength=0x14001000 > > HOST_TO_IOC > > SE32 0xe653dc90: Addr=0x6f09000 FlagsLength=0x14001000 > > HOST_TO_IOC > > SE32 0xe653dc98: Addr=0x4c2aa000 FlagsLength=0x94001000 > > HOST_TO_IOC LAST_ELEMENT > > CE32 0xe653dca0: Addr=0x7d610ca8 NxtChnO=0x0 Flgs=0x30 Len=0x18 > > SE32 0xe653dca8: Addr=0x3294b000 FlagsLength=0x14001000 > > HOST_TO_IOC > > SE32 0xe653dcb0: Addr=0x6e2ec000 FlagsLength=0x14001000 > > HOST_TO_IOC > > SE32 0xe653dcb8: Addr=0x2d78d000 FlagsLength=0xd5001000 > > HOST_TO_IOC LAST_ELEMENT END_OF_BUFFER END_OF_LIST > > mpt0: time out on request index = 0x36 sequence = 0x00246949 > > mpt0: Status 00000001; Mask 00000001; Doorbell 24000000 > > request state On Chip > > SCSI IO Request @ 0xe6489c20 > > Chain Offset 0x10 > > MsgFlags 0x00 > > MsgContext 0x00000036 > > Bus: 0 > > TargetID 0 > > SenseBufferLength 32 > > LUN: 0x0 > > Control 0x01000000 WRITE SIMPLEQ > > DataLength 0x00010000 > > SenseBufAddr 0x7d5f8de0 > > CDB[0:10] 2a 00 6c 59 d7 d0 00 00 80 00 > > SE32 0xe6525c30: Addr=0x878e000 FlagsLength=0x14001000 > > HOST_TO_IOC > > SE32 0xe6525c38: Addr=0x7906f000 FlagsLength=0x94001000 > > HOST_TO_IOC LAST_ELEMENT > > CE32 0xe6525c40: Addr=0x7d5f8c48 NxtChnO=0x16 Flgs=0x30 Len=0x60 > > SE32 0xe6525c48: Addr=0x21170000 FlagsLength=0x14001000 > > HOST_TO_IOC > > SE32 0xe6525c50: Addr=0x503f1000 FlagsLength=0x14001000 > > HOST_TO_IOC > > SE32 0xe6525c58: Addr=0x58c92000 FlagsLength=0x14001000 > > HOST_TO_IOC > > SE32 0xe6525c60: Addr=0x29a33000 FlagsLength=0x14001000 > > HOST_TO_IOC > > SE32 0xe6525c68: Addr=0x181d4000 FlagsLength=0x14001000 > > HOST_TO_IOC > > SE32 0xe6525c70: Addr=0x2875000 FlagsLength=0x14001000 > > HOST_TO_IOC > > SE32 0xe6525c78: Addr=0x58c76000 FlagsLength=0x14001000 > > HOST_TO_IOC > > SE32 0xe6525c80: Addr=0x4937000 FlagsLength=0x14001000 > > HOST_TO_IOC > > SE32 0xe6525c88: Addr=0x5fc78000 FlagsLength=0x14001000 > > HOST_TO_IOC > > SE32 0xe6525c90: Addr=0x7d019000 FlagsLength=0x14001000 > > HOST_TO_IOC > > SE32 0xe6525c98: Addr=0x7915a000 FlagsLength=0x94001000 > > HOST_TO_IOC LAST_ELEMENT > > CE32 0xe6525ca0: Addr=0x7d5f8ca8 NxtChnO=0x0 Flgs=0x30 Len=0x18 > > SE32 0xe6525ca8: Addr=0x909b000 FlagsLength=0x14001000 > > HOST_TO_IOC > > SE32 0xe6525cb0: Addr=0x27f9c000 FlagsLength=0x14001000 > > HOST_TO_IOC > > SE32 0xe6525cb8: Addr=0x65a1d000 FlagsLength=0xd5001000 > > HOST_TO_IOC LAST_ELEMENT END_OF_BUFFER END_OF_LIST > > Cheers, > -- > Xin LI http://www.delphij.net/