From owner-freebsd-stable@FreeBSD.ORG Sat Jan 26 01:21:24 2008 Return-Path: Delivered-To: freebsd-stable@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id B5EE316A46B for ; Sat, 26 Jan 2008 01:21:24 +0000 (UTC) (envelope-from jdc@parodius.com) Received: from mx01.sc1.parodius.com (mx01.sc1.parodius.com [72.20.106.3]) by mx1.freebsd.org (Postfix) with ESMTP id 9104313C469 for ; Sat, 26 Jan 2008 01:21:24 +0000 (UTC) (envelope-from jdc@parodius.com) Received: by mx01.sc1.parodius.com (Postfix, from userid 1000) id 6D61E1CC05F; Fri, 25 Jan 2008 17:21:24 -0800 (PST) Date: Fri, 25 Jan 2008 17:21:24 -0800 From: Jeremy Chadwick To: Joe Peterson Message-ID: <20080126012124.GA53400@eos.sc1.parodius.com> References: <479A0731.6020405@skyrush.com> <20080125162940.GA38494@eos.sc1.parodius.com> <479A3764.6050800@skyrush.com> <3803988D-8D18-4E89-92EA-19BF62FD2395@mac.com> <479A4CB0.5080206@skyrush.com> <20080126003845.GA52183@eos.sc1.parodius.com> <479A86E5.5060806@skyrush.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <479A86E5.5060806@skyrush.com> User-Agent: Mutt/1.5.16 (2007-06-09) Cc: freebsd-stable@freebsd.org Subject: Re: "ad0: TIMEOUT - WRITE_DMA" type errors with 7.0-RC1 X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 26 Jan 2008 01:21:24 -0000 On Fri, Jan 25, 2008 at 06:03:33PM -0700, Joe Peterson wrote: > Wow, pretty crazy! Hmm, and yes, those LBAs do look close together. > Well, let me know how the smartctl output looks. I'd be curious if your > bad sector count rises. Absolutely nada on the SMART statistics. Nothing incremented or changed in any way. My short and long tests did not change any of the data in the fields either. Full output is below my .sig. > BTW, I tried: > > crater# dd if=/dev/ad1s4 of=/dev/null bs=64k > ^C1408596+0 records in > 1408596+0 records out > 92313747456 bytes transferred in 1415.324362 secs (65224446 bytes/sec) > > (I let it go for 92GB or so) - no messages about ad1. So I wonder if > this points at either the cable connector on ad0 or the drive itself. I > guess I'd rather have a failing drive than motherboard... > > I originally was wondering if somehow something peculiar about ZFS's > disk access pattern was making it happen... Since I'm used to dealing with disk issues (at work and personally), I'm left wondering if this is some strange ATA subsystem quirk, or ultimately something with ZFS (your "something peculiar about ZFS's disk access pattern" claim is starting to look more plausible). This may sound suicidal, but I'm hoping to recreate the scenario somehow, and then punt the details to Soren or Xin Li for further investigation -- if it looks like an ATA subsystem thing, that is. It is interesting to note that we both have Seagate disks... :-) I'll have to run SeaTools on my disk to see if anything comes back, or run a selective LBA test in smartctl (since the drive supports it). I've restarted my rsync since, and it's happily chomping away without an issue. If my problem was TRULY a bad block or something causing mechanical lock-up on the disk, I'd have expected my latest rsync to induce it. There's always the chance of some bizarre drive firmware bug too. > THanks for the recomendations. I'll keep an eye on it, and I'll let you > know what a cable change does for me. Still, I have not had any ad0 > messages since this morning (I haven't been using the system today much, > but maybe the cron processes are more likely to trigger it... Understood. In my case, I *know* the cables are fine, because the box itself I just built and migrated to a few days ago (change of motherboard, chassis, and addition of SATA hot-swap backplane). We use the same motherboard (Supermicro PDSMI+) in all of our production servers in our datacenter, and they're rock-solid. I've done hot-swapping without any issue on those systems too, and I've never seen any SATA system issues -- one of the systems is our datacenter backup server, which holds nightly backups for all the other boxes (about 6). Due to the heavy disk I/O that occurs for hours at a time, if this was some weird system quirk, motherboard problem, or SATA bus/cable issue, we would've seen it by now. FWIW: all our systems, including the backup box, use UFS2 exclusively -- no ZFS in the picture. -- | Jeremy Chadwick jdc at parodius.com | | Parodius Networking http://www.parodius.com/ | | UNIX Systems Administrator Mountain View, CA, USA | | Making life hard for others since 1977. PGP: 4BD6C0CB | icarus# smartctl -a /dev/ad6 smartctl version 5.37 [i386-portbld-freebsd7.0] Copyright (C) 2002-6 Bruce Allen Home page is http://smartmontools.sourceforge.net/ === START OF INFORMATION SECTION === Model Family: Seagate Barracuda 7200.10 family Device Model: ST3500630AS Serial Number: 9QG1YWNL Firmware Version: 3.AAE User Capacity: 500,107,862,016 bytes Device is: In smartctl database [for details use: -P show] ATA Version is: 7 ATA Standard is: Exact ATA specification draft version not indicated Local Time is: Fri Jan 25 17:10:31 2008 PST SMART support is: Available - device has SMART capability. SMART support is: Enabled === START OF READ SMART DATA SECTION === SMART overall-health self-assessment test result: PASSED General SMART Values: Offline data collection status: (0x82) Offline data collection activity was completed without error. Auto Offline Data Collection: Enabled. Self-test execution status: ( 0) The previous self-test routine completed without error or no self-test has ever been run. Total time to complete Offline data collection: ( 430) seconds. Offline data collection capabilities: (0x5b) SMART execute Offline immediate. Auto Offline data collection on/off support. Suspend Offline collection upon new command. Offline surface scan supported. Self-test supported. No Conveyance Self-test supported. Selective Self-test supported. SMART capabilities: (0x0003) Saves SMART data before entering power-saving mode. Supports SMART auto save timer. Error logging capability: (0x01) Error logging supported. General Purpose Logging supported. Short self-test routine recommended polling time: ( 1) minutes. Extended self-test routine recommended polling time: ( 163) minutes. SMART Attributes Data Structure revision number: 10 Vendor Specific SMART Attributes with Thresholds: ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE 1 Raw_Read_Error_Rate 0x000f 114 094 006 Pre-fail Always - 131599973 3 Spin_Up_Time 0x0003 094 094 000 Pre-fail Always - 0 4 Start_Stop_Count 0x0032 100 100 020 Old_age Always - 6 5 Reallocated_Sector_Ct 0x0033 100 100 036 Pre-fail Always - 0 7 Seek_Error_Rate 0x000f 082 060 030 Pre-fail Always - 200325271 9 Power_On_Hours 0x0032 097 097 000 Old_age Always - 2970 10 Spin_Retry_Count 0x0013 100 100 097 Pre-fail Always - 0 12 Power_Cycle_Count 0x0032 100 100 020 Old_age Always - 9 187 Unknown_Attribute 0x0032 100 100 000 Old_age Always - 0 189 Unknown_Attribute 0x003a 100 100 000 Old_age Always - 0 190 Temperature_Celsius 0x0022 063 050 045 Old_age Always - 773849125 194 Temperature_Celsius 0x0022 037 050 000 Old_age Always - 37 (Lifetime Min/Max 0/29) 195 Hardware_ECC_Recovered 0x001a 062 059 000 Old_age Always - 5882952 197 Current_Pending_Sector 0x0012 100 100 000 Old_age Always - 0 198 Offline_Uncorrectable 0x0010 100 100 000 Old_age Offline - 0 199 UDMA_CRC_Error_Count 0x003e 200 200 000 Old_age Always - 0 200 Multi_Zone_Error_Rate 0x0000 100 253 000 Old_age Offline - 0 202 TA_Increase_Count 0x0032 100 253 000 Old_age Always - 0 SMART Error Log Version: 1 No Errors Logged SMART Self-test log structure revision number 1 Num Test_Description Status Remaining LifeTime(hours) LBA_of_first_error # 1 Short offline Completed without error 00% 2970 - # 2 Short offline Completed without error 00% 321 - SMART Selective self-test log data structure revision number 1 SPAN MIN_LBA MAX_LBA CURRENT_TEST_STATUS 1 0 0 Not_testing 2 0 0 Not_testing 3 0 0 Not_testing 4 0 0 Not_testing 5 0 0 Not_testing Selective self-test flags (0x0): After scanning selected spans, do NOT read-scan remainder of disk. If Selective self-test is pending on power-up, resume after 0 minute delay.