larson

Thanks, JorgeB, that confirms my research. Out of curiosity, is there a limit to the rootfs when mounted, or will it just use up all available memory? I have managed to stop the Deluge docker, and moved away the files from that path. Will have to reboot from command line now, since the webgui didn't like to be lacking RAM.

Running on root file system in RAM? rootfs on /mnt type rootfs (rw,size=98752492k,nr_inodes=24688123,inode64) That would be the next step up in mounted FS in the system, wouldn't it? Haven't ever had a reason to consider what a rootfs is. Guess I'll know soon... Lars Olof

Maybe not properly related to this group, but will start here. Just moved my stuff from old rig. New rig was set upp with disk 5-12 mapped during trials. Old rig had 1-12 or so. On transfer of license, I set up new rig with 1-8 mapped (still no parity active). When restoring dockers from Appdata Backup I didn't think long enough, and just left the previous mapping of my Deluge container /downloads > /mnt/disk9/noshare/deluge/ and started the container. Everything runs fine, downloads being done and copying finished to /mnt/disk9/noshare/deluge/done, as expected. I can see from CLI /mnt/Disk9... etc, and copy out files from there. But since I haven't yet assigned a Disk9, it is not listed in the Main array list. Now I am just curious and worried of what is actually backing these files. Are they saved inside the container, or on the flash? Maybe the cache drive with appdata folder? I checked with lsblk, but no sign of it there (not even / mount point listed) Any ideas, or suggestions where to check? I'll see if current DLs can finish, and stop to reconfigure after that. Keeping an eye on the disk utilisations of flash, docker image and cache until then. I can provide screens or CLI output, if someone would find it helpful. Lars Olof

Thank you, Vr, I was actually thinking I'd keep using the old USB (flash) to keep the licence unchanged. The USB isn't used much during Unraid operation, and should not have any wear or fatigue related issues, am I right? To follow your suggestion, I would first have to get a reputable flash drive for the new server, currently using some old unknown that was lying around at home. Lars Olof

TL/DR Best way to move from old server to new server, with unsimilar hardware layout I have a old server that I want to replace with new hardware. Got hold of new hardware this spring (HP gen10 server with HP SAS RAID), requested trial license to a new USB, and got it up and running. Only annoying part of it is SAS drives seem to not spin down. Can live with that. Using containers for Plex, Nextcloud and Deluge, plus some less important ones. No VMs, but a few plugins. On new server with trial license, precleared 10 drives, formatted and got things accessible. Didn't configure parity. Moved 25TB data from old server with rsync to new drives. Tried to keep to old top folder structure on drives. (music/TV/Film in root) Then time was up, both on license and my available time (started move from apt to house, no time for messing with servers) Old user shares was spread across drives, have tried to keep them on single drives on new HW. Today I'm still running old server, with changed data (after rsyncing) in Deluge DL folder and possibly Plex metadata and Nextcloud. Just updated old server to 6.12 (I believe that was what was installed on new server also, it's currently offline). Have installed Appdata Backup plugin, and made backups of my dockers. Planning to rsync the changes in Deluge and Nextcloud data folders before switching. What is the right way to move my old flash (in the license) to the new server, without to much problems in getting up and running again. I have a flash backup by Appdata available, should I just save my key, wipe the flash and reinstall Unraid 6.latest on it, and start from fresh with my old key, restore appdata for my dockers, share selected folders and be on my way? Manually reinstall my plugins? Any suggestions, known pitfalls, convenient shortcuts? Thanks in advance for any shared wisdom. Lars Olof

To put some answers here for someone searching in the future. I have successfully begun transferring data to the server. The on-board (or rather proprietary socket card) SA P816i-a raid adapter runs fine in Unraid. When the drives is not configured in an array, they are automatically set for HBA mode. Just start using them in Unraid. Minor trouble remaining, spindown and possibly some SMART features are not running correctly on SAS drives. As expected. Found a bunch of SATA interfaces on the board, but no free power interfaces. One SATA was mini-SATA combo power and data (supposedly for a DVD). Did a MacGyver, ordered a 2x2.5" in 3.5" bracket, a couple of adapters and split the power from the mini-SATA to both drives, and used another data only port for my two SSD. Taped the bracket to the bottom of the case, lets me use 6" cables, keeping it neat. Seems to work fine so far. i have precleared the 12 4TB drives that came with the machine, replaced a couple with one SAS and one SATA 8TB drive. When preclearing the 8TB drives, the SATA ran 10% shorter time. That will be my parity drive. Have currently created an array with one 8 TB and 7 4 TB drives without parity and have started transfering data from the old server. Purchased a HP 562 2x10Gbit SFP+ card for the old server, using a DAC. Set up the 10G interfaces on both servers in their own IP net, and direct connected the NICs. Copying files with rsync initiated on the new server, Three transfers simultaneously, from and to separate disks (using /mnt/diskx/folder/ path) Total network load, about 2 Gbits on the link. 90% cpu over 4 cores on old server. 10-20% load over 16 vcores on the new server. Looking good. Planning on transfering everything over network, to much trouble trying to move disks.

If you have only the failed 3 and the parity 12, and want to keep the data from the 3, I believe you'll have to either replace the 3 with a new 3-12 disk or transfer data from the unprotected array to another location (network, unassigned device) and rebuild your array later. If you have more disks in the array with enough free space, you could move data from the failed disk (parity calculated) to the working disks. Probably pretty slow, with high parity utilization. But with no other device smaller than 12, you'd have to break the array and rebuild it anyhow, Lacking a smaller (3-12) drive, I would probably move data from failed disk to other disks in array, break array, remove 3, reuse 12 as array disk and add 14 as parity. Then purchase new 14 as spare, or possibly a pair of large disks at future opportunity to replace parity and spare disk. Lars Olof

As in an earlier post, I am setting up a new server, based on a decommissioned Proliant G10. Came with a full rack (4 bays of 4 3.5" SAS backplanes) stacked with HP 4 TB disks. All connected to the original HP Smart Array P816i-a SR Gen10. Obviously unconfiguring the array is enough to present disks in HBA mode for this card. In addition I have some 8 TB disks, a Seagate SAS Enterprise Capacity ST8000NM0075 and a Seagate SATA Exos 7E10 ST8000NM017B-2TJ (and another similar SATA running as parity in my current system, to be migrated later). There seem to be a slight performance advantage for the SATA drive, it ran the preclear cycle in 31h52m compared to 35h41m for the SAS drive. My future plan is to run only SATA on one backplane, and keep SAS only on the other two. In your opinion, which disk should I use as parity? Is there any advantage in using a SAS drive, that can counter the slower apparent speed compared to the SATA? Is it any advantage to keep the different kind of disks on separate backplanes?

As suggested in topic, I have gotten hold of a more recent machine to remake into Unraid server. This one arrives with 12x4TB SAS 3,5" drives in 3 bays, on the original SmartArray P816i-a SR Gen10. From what I have found, it seems drives on that card will run in HBA mode if not configured to an array. Sounds fine to me. If I wanted to keep sane, I'd probably just install Unraid, and be done with it. I don't want to keep sane In my current setup, I have a 8 TB SATA for parity, and another that I haven't installed yet. Other drives are 2, 3 and 4 TB for a total of 11 data drives. I also have a 128GB SATA SSD cache installed. Would I gain performance by purchasing a new SAS 8 TB disk for parity, instead of reusing my current 8's? Would I gain performance by running all disks in a bay (SAS backplane) on SAS, and not mix SAS and SATA on the same channel on the controller? Should I install my new SSD cache disks (2x480GB SATA 2,5") on the current RAID card, or maybe rather on (probable) onboard SATA ports? I believe there is a fourth SAS channel on the card, and I got breakouts to connect them outside of the bays, to keep it homegrown. If it is not recommended to use the SA P816i-a or onboard SATA, should I try to reuse controllers from my current build? System devices tells me I have an Adaptec 1430 SA (unknown number of ports, 4?), Highpoint Rocketraid 230x 4 port SATA II and a LSI SAS2008 SAS-2 (Probably an HBA:d IBM Serverraid, 2 or more SAS connectors) My vaugeness here is from buying and installing this stuff a long time ago, and just checking Unraid system devices now to see what I got. Should I try to get hold of a 10Gbit ethernet card for my old server, to transfer data to my new server, the new has one installed already. I have bad experiences with mounting drives outside of the array to copy files, I'd rather set up an empty array and copy (Rsync?) the files across. What could be good value for money, and good track record with SFP(+) ports, I can find HP 552 and 562 and others at my local used hardware dealer. Well, many questions, I'm hoping for some initiated answers. I'd rather be stopped before I start doing stupid things, than to discover them after doing all the hard work. /Lars Olof Norell

And the most probable reason for this would be a bad disk? Or might it be controller or cable? I'll be running it tonight in an internal bay to check it out, more results in about 12 hours. [Edit] Didn't need any 12 hours to get a result. When running in an internal slot the results are similar to the results in the ESATA cradle. Is there anything in the SMART reports that I can use when returning the disk? Latest one is here: Disk: /dev/sdk smartctl 5.40 2010-10-16 r3189 [i486-slackware-linux-gnu] (local build) Copyright (C) 2002-10 by Bruce Allen, http://smartmontools.sourceforge.net === START OF INFORMATION SECTION === Device Model: WDC WD30EZRX-00MMMB0 Serial Number: WD-WCAWZ1377700 Firmware Version: 80.00A80 User Capacity: 3,000,592,982,016 bytes Device is: Not in smartctl database [for details use: -P showall] ATA Version is: 8 ATA Standard is: Exact ATA specification draft version not indicated Local Time is: Fri Nov 18 20:28:14 2011 CET 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: (0x80) Offline data collection activity was never started. 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: (49680) seconds. Offline data collection capabilities: (0x7b) SMART execute Offline immediate. Auto Offline data collection on/off support. Suspend Offline collection upon new command. Offline surface scan supported. Self-test supported. 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: ( 2) minutes. Extended self-test routine recommended polling time: ( 255) minutes. Conveyance self-test routine recommended polling time: ( 5) minutes. SCT capabilities: (0x3035) SCT Status supported. SCT Feature Control supported. SCT Data Table supported. SMART Attributes Data Structure revision number: 16 Vendor Specific SMART Attributes with Thresholds: ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE 1 Raw_Read_Error_Rate 0x002f 188 188 051 Pre-fail Always - 133 3 Spin_Up_Time 0x0027 192 178 021 Pre-fail Always - 7391 4 Start_Stop_Count 0x0032 100 100 000 Old_age Always - 13 5 Reallocated_Sector_Ct 0x0033 200 200 140 Pre-fail Always - 2 7 Seek_Error_Rate 0x002e 200 200 000 Old_age Always - 0 9 Power_On_Hours 0x0032 100 100 000 Old_age Always - 36 10 Spin_Retry_Count 0x0032 100 253 000 Old_age Always - 0 11 Calibration_Retry_Count 0x0032 100 253 000 Old_age Always - 0 12 Power_Cycle_Count 0x0032 100 100 000 Old_age Always - 10 192 Power-Off_Retract_Count 0x0032 200 200 000 Old_age Always - 8 193 Load_Cycle_Count 0x0032 200 200 000 Old_age Always - 33 194 Temperature_Celsius 0x0022 116 108 000 Old_age Always - 36 196 Reallocated_Event_Count 0x0032 199 199 000 Old_age Always - 1 197 Current_Pending_Sector 0x0032 200 200 000 Old_age Always - 3 198 Offline_Uncorrectable 0x0030 100 253 000 Old_age Offline - 0 199 UDMA_CRC_Error_Count 0x0032 200 200 000 Old_age Always - 0 200 Multi_Zone_Error_Rate 0x0008 100 253 000 Old_age Offline - 0 SMART Error Log Version: 1 ATA Error Count: 6686 (device log contains only the most recent five errors) CR = Command Register [HEX] FR = Features Register [HEX] SC = Sector Count Register [HEX] SN = Sector Number Register [HEX] CL = Cylinder Low Register [HEX] CH = Cylinder High Register [HEX] DH = Device/Head Register [HEX] DC = Device Command Register [HEX] ER = Error register [HEX] ST = Status register [HEX] Powered_Up_Time is measured from power on, and printed as DDd+hh:mm:SS.sss where DD=days, hh=hours, mm=minutes, SS=sec, and sss=millisec. It "wraps" after 49.710 days. Error 6686 occurred at disk power-on lifetime: 36 hours (1 days + 12 hours) When the command that caused the error occurred, the device was active or idle. After command completion occurred, registers were: ER ST SC SN CL CH DH -- -- -- -- -- -- -- 01 51 00 90 ac ff e2 Error: AMNF at LBA = 0x02ffac90 = 50310288 Commands leading to the command that caused the error were: CR FR SC SN CL CH DH DC Powered_Up_Time Command/Feature_Name -- -- -- -- -- -- -- -- ---------------- -------------------- c8 00 00 b8 ab ff e2 08 00:45:51.078 READ DMA ec 00 00 00 00 00 a0 08 00:45:51.058 IDENTIFY DEVICE ef 03 45 00 00 00 a0 08 00:45:51.058 SET FEATURES [set transfer mode] Error 6685 occurred at disk power-on lifetime: 36 hours (1 days + 12 hours) When the command that caused the error occurred, the device was active or idle. After command completion occurred, registers were: ER ST SC SN CL CH DH -- -- -- -- -- -- -- 01 51 00 90 ac ff e2 Error: AMNF at LBA = 0x02ffac90 = 50310288 Commands leading to the command that caused the error were: CR FR SC SN CL CH DH DC Powered_Up_Time Command/Feature_Name -- -- -- -- -- -- -- -- ---------------- -------------------- c8 00 00 b8 ab ff e2 08 00:45:42.359 READ DMA c8 00 00 b8 aa ff e2 08 00:45:42.358 READ DMA c8 00 00 b8 a9 ff e2 08 00:45:42.357 READ DMA c8 00 00 b8 a8 ff e2 08 00:45:42.356 READ DMA c8 00 00 b8 a7 ff e2 08 00:45:42.355 READ DMA Error 6684 occurred at disk power-on lifetime: 36 hours (1 days + 12 hours) When the command that caused the error occurred, the device was active or idle. After command completion occurred, registers were: ER ST SC SN CL CH DH -- -- -- -- -- -- -- 01 51 00 98 1d d4 e2 Error: AMNF at LBA = 0x02d41d98 = 47455640 Commands leading to the command that caused the error were: CR FR SC SN CL CH DH DC Powered_Up_Time Command/Feature_Name -- -- -- -- -- -- -- -- ---------------- -------------------- c8 00 00 f0 1c d4 e2 08 00:45:00.946 READ DMA ec 00 00 00 00 00 a0 08 00:45:00.927 IDENTIFY DEVICE ef 03 46 00 00 00 a0 08 00:45:00.927 SET FEATURES [set transfer mode] Error 6683 occurred at disk power-on lifetime: 36 hours (1 days + 12 hours) When the command that caused the error occurred, the device was active or idle. After command completion occurred, registers were: ER ST SC SN CL CH DH -- -- -- -- -- -- -- 40 51 00 98 1d d4 e2 Error: UNC at LBA = 0x02d41d98 = 47455640 Commands leading to the command that caused the error were: CR FR SC SN CL CH DH DC Powered_Up_Time Command/Feature_Name -- -- -- -- -- -- -- -- ---------------- -------------------- c8 00 00 f0 1c d4 e2 08 00:44:52.477 READ DMA c8 00 00 f0 1b d4 e2 08 00:44:52.476 READ DMA c8 00 00 f0 1a d4 e2 08 00:44:52.475 READ DMA c8 00 00 f0 19 d4 e2 08 00:44:52.473 READ DMA c8 00 00 f0 18 d4 e2 08 00:44:52.472 READ DMA Error 6682 occurred at disk power-on lifetime: 36 hours (1 days + 12 hours) When the command that caused the error occurred, the device was active or idle. After command completion occurred, registers were: ER ST SC SN CL CH DH -- -- -- -- -- -- -- 40 51 00 28 04 d4 e2 Error: UNC at LBA = 0x02d40428 = 47449128 Commands leading to the command that caused the error were: CR FR SC SN CL CH DH DC Powered_Up_Time Command/Feature_Name -- -- -- -- -- -- -- -- ---------------- -------------------- c8 00 00 f0 03 d4 e2 08 00:44:37.369 READ DMA ec 00 00 00 00 00 a0 08 00:44:37.349 IDENTIFY DEVICE ef 03 46 00 00 00 a0 08 00:44:37.349 SET FEATURES [set transfer mode] SMART Self-test log structure revision number 1 No self-tests have been logged. [To run self-tests, use: smartctl -t] 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. /Lars Olof

I have tried again to run Preclear on this disk. Since I run it in a external cradle (similar to this: http://www.storagedepot.co.uk/External-Hard-Drives/Hard-Drive-Docks/sc883/p753popup1532.aspx) I cannot do much to the connectors. I relocated the cradle, reseated the disk and checked the cabling. But still the run is crawling to a halt. 2% overnight :-( I'll see if I can find a free internal sata connector to run it there. For now I give you logs from this run. What I would like to know is if the lack of performance might be a result of bad power, or if the interface or even the disk is giving up. First just the SMART snippets, first report: ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE 1 Raw_Read_Error_Rate 0x002f 193 191 051 Pre-fail Always - 24 3 Spin_Up_Time 0x0027 196 178 021 Pre-fail Always - 7166 4 Start_Stop_Count 0x0032 100 100 000 Old_age Always - 12 5 Reallocated_Sector_Ct 0x0033 200 200 140 Pre-fail Always - 1 7 Seek_Error_Rate 0x002e 100 253 000 Old_age Always - 0 9 Power_On_Hours 0x0032 100 100 000 Old_age Always - 29 10 Spin_Retry_Count 0x0032 100 253 000 Old_age Always - 0 11 Calibration_Retry_Count 0x0032 100 253 000 Old_age Always - 0 12 Power_Cycle_Count 0x0032 100 100 000 Old_age Always - 9 192 Power-Off_Retract_Count 0x0032 200 200 000 Old_age Always - 7 193 Load_Cycle_Count 0x0032 200 200 000 Old_age Always - 27 194 Temperature_Celsius 0x0022 112 108 000 Old_age Always - 40 196 Reallocated_Event_Count 0x0032 200 200 000 Old_age Always - 0 197 Current_Pending_Sector 0x0032 200 200 000 Old_age Always - 3 198 Offline_Uncorrectable 0x0030 100 253 000 Old_age Offline - 0 199 UDMA_CRC_Error_Count 0x0032 200 200 000 Old_age Always - 0 200 Multi_Zone_Error_Rate 0x0008 100 253 000 Old_age Offline - 0 Then rerun after breaking after a few hours: ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE 1 Raw_Read_Error_Rate 0x002f 194 191 051 Pre-fail Always - 48 3 Spin_Up_Time 0x0027 196 178 021 Pre-fail Always - 7166 4 Start_Stop_Count 0x0032 100 100 000 Old_age Always - 12 5 Reallocated_Sector_Ct 0x0033 200 200 140 Pre-fail Always - 2 7 Seek_Error_Rate 0x002e 200 200 000 Old_age Always - 0 9 Power_On_Hours 0x0032 100 100 000 Old_age Always - 35 10 Spin_Retry_Count 0x0032 100 253 000 Old_age Always - 0 11 Calibration_Retry_Count 0x0032 100 253 000 Old_age Always - 0 12 Power_Cycle_Count 0x0032 100 100 000 Old_age Always - 9 192 Power-Off_Retract_Count 0x0032 200 200 000 Old_age Always - 7 193 Load_Cycle_Count 0x0032 200 200 000 Old_age Always - 29 194 Temperature_Celsius 0x0022 108 108 000 Old_age Always - 44 196 Reallocated_Event_Count 0x0032 199 199 000 Old_age Always - 1 197 Current_Pending_Sector 0x0032 200 200 000 Old_age Always - 3 198 Offline_Uncorrectable 0x0030 100 253 000 Old_age Offline - 0 199 UDMA_CRC_Error_Count 0x0032 200 200 000 Old_age Always - 0 200 Multi_Zone_Error_Rate 0x0008 100 253 000 Old_age Offline - 0 I would also attach my syslogs (it looks to have split tonight), the dmesg and the two full preclear start logs, but the zip gets too big. Hope it will work from here: http://www.larsolof.se/files/syslogs.zip My power issue looks fixed, since the value hasn't increased on this run. Something else is no good. /Lars Olof

Current SMART status from Unmenu: SMART Attributes Data Structure revision number: 16 Vendor Specific SMART Attributes with Thresholds: ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE 1 Raw_Read_Error_Rate 0x002f 193 191 051 Pre-fail Always - 24 3 Spin_Up_Time 0x0027 202 178 021 Pre-fail Always - 6900 4 Start_Stop_Count 0x0032 100 100 000 Old_age Always - 11 5 Reallocated_Sector_Ct 0x0033 200 200 140 Pre-fail Always - 1 7 Seek_Error_Rate 0x002e 100 253 000 Old_age Always - 0 9 Power_On_Hours 0x0032 100 100 000 Old_age Always - 10 10 Spin_Retry_Count 0x0032 100 253 000 Old_age Always - 0 11 Calibration_Retry_Count 0x0032 100 253 000 Old_age Always - 0 12 Power_Cycle_Count 0x0032 100 100 000 Old_age Always - 8 192 Power-Off_Retract_Count 0x0032 200 200 000 Old_age Always - 7 193 Load_Cycle_Count 0x0032 200 200 000 Old_age Always - 19 194 Temperature_Celsius 0x0022 113 108 000 Old_age Always - 39 196 Reallocated_Event_Count 0x0032 200 200 000 Old_age Always - 0 197 Current_Pending_Sector 0x0032 200 200 000 Old_age Always - 3 198 Offline_Uncorrectable 0x0030 100 253 000 Old_age Offline - 0 199 UDMA_CRC_Error_Count 0x0032 200 200 000 Old_age Always - 0 200 Multi_Zone_Error_Rate 0x0008 100 253 000 Old_age Offline - 0 Status of disk is started once in ESATA cradle, started running Preclear, aborted it, spin down once and left in cradle overnight. Makes me curious about "12 Power_Cycle_Count 0x0032 100 100 000 Old_age Always - 8" /Lars Olof

Ok, so I got a 3TB and started a preclear run. After an hour or so I checked in on it, seeing this: ================================================================== 1.12 = unRAID server Pre-Clear disk /dev/sda = cycle 1 of 1, partition start on sector 1 = Disk Pre-Read in progress: 0% complete = ( 27,965,952,000 bytes of 3,000,592,982,016 read ) 1.3 MB/s = = = = = = = = = = Disk Temperature: 43C, Elapsed Time: 1:01:02 Doesn't look too good. It's running hot, but most of all at snails pace. Well, a look in the syslog shows something is rotten in the state of Denmark: Nov 16 21:47:29 Tower kernel: ata4.00: BMDMA stat 0x64 Nov 16 21:47:29 Tower kernel: ata4.00: failed command: READ DMA Nov 16 21:47:29 Tower kernel: ata4.00: cmd c8/00:08:d8:41:65/00:00:00:00:00/e3 tag 0 dma 4096 in Nov 16 21:47:29 Tower kernel: res 51/40:08:d8:41:65/00:00:00:00:00/e3 Emask 0x9 (media error) Nov 16 21:47:29 Tower kernel: ata4.00: status: { DRDY ERR } Nov 16 21:47:29 Tower kernel: ata4.00: error: { UNC } Nov 16 21:47:29 Tower kernel: ata4.00: configured for UDMA/133 Nov 16 21:47:29 Tower kernel: ata4: EH complete Nov 16 21:47:31 Tower kernel: ata4.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x0 Nov 16 21:47:31 Tower kernel: ata4.00: BMDMA stat 0x64 Nov 16 21:47:31 Tower kernel: ata4.00: failed command: READ DMA Nov 16 21:47:31 Tower kernel: ata4.00: cmd c8/00:08:d8:41:65/00:00:00:00:00/e3 tag 0 dma 4096 in Nov 16 21:47:31 Tower kernel: res 51/40:08:d8:41:65/00:00:00:00:00/e3 Emask 0x9 (media error) Nov 16 21:47:31 Tower kernel: ata4.00: status: { DRDY ERR } Nov 16 21:47:31 Tower kernel: ata4.00: error: { UNC } Nov 16 21:47:31 Tower kernel: ata4.00: configured for UDMA/133 Nov 16 21:47:31 Tower kernel: sd 5:0:0:0: [sda] Unhandled sense code Nov 16 21:47:31 Tower kernel: sd 5:0:0:0: [sda] Result: hostbyte=0x00 driverbyte=0x08 Nov 16 21:47:31 Tower kernel: sd 5:0:0:0: [sda] Sense Key : 0x3 [current] [descriptor] Nov 16 21:47:31 Tower kernel: Descriptor sense data with sense descriptors (in hex): Nov 16 21:47:31 Tower kernel: 72 03 11 04 00 00 00 0c 00 0a 80 00 00 00 00 00 Nov 16 21:47:31 Tower kernel: 03 65 41 d8 Nov 16 21:47:31 Tower kernel: sd 5:0:0:0: [sda] ASC=0x11 ASCQ=0x4 Nov 16 21:47:31 Tower kernel: sd 5:0:0:0: [sda] CDB: cdb[0]=0x28: 28 00 03 65 41 d8 00 00 08 00 Nov 16 21:47:31 Tower kernel: end_request: I/O error, dev sda, sector 56967640 Nov 16 21:47:31 Tower kernel: Buffer I/O error on device sda, logical block 7120955 Nov 16 21:47:31 Tower kernel: ata4: EH complete Nov 16 21:47:34 Tower kernel: ata4.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x0 Nov 16 21:47:34 Tower kernel: ata4.00: BMDMA stat 0x64 Nov 16 21:47:34 Tower kernel: ata4.00: failed command: READ DMA Nov 16 21:47:34 Tower kernel: ata4.00: cmd c8/00:08:28:7d:65/00:00:00:00:00/e3 tag 0 dma 4096 in Nov 16 21:47:34 Tower kernel: res 51/40:08:28:7d:65/00:00:00:00:00/e3 Emask 0x9 (media error) Nov 16 21:47:34 Tower kernel: ata4.00: status: { DRDY ERR } Nov 16 21:47:34 Tower kernel: ata4.00: error: { UNC } Nov 16 21:47:34 Tower kernel: ata4.00: configured for UDMA/133 Nov 16 21:47:34 Tower kernel: ata4: EH complete And this just repeats over and over. Setup is an ESATA cradle on a marvell onboard chip. I haven't noticed any particular performance problems with it before, but will now relocate it (temp prob) and check the cabling. Is it possible to see any reason from the logs above, or other logs, or should I just trial and error to find a working setup (or broken disk). All of my internal SATA is busy with the array, so I cannot try it there. /Lars Olof

Ok, has run preclear a couple of times on one of my drives now. I have ran it one cycle at a time. (BTW, Joe, it's great logging the results on the flash share, but your naming convention seem to nuke results coming in the same date. Not a big problem with the 2 TB disks that take around 24h to run, but I think I lost a few results on my 12 h runs of this 1TB disk.) Now I have 3 results from these preclears, and there is a nagging feeling not everything is right. ============================================================================ ** Changed attributes in files: /tmp/smart_start_sdb /tmp/smart_finish_sdb ATTRIBUTE NEW_VAL OLD_VAL FAILURE_THRESHOLD STATUS RAW_VALUE Raw_Read_Error_Rate = 118 112 6 ok 174433234 Spin_Retry_Count = 100 100 97 near_thresh 0 End-to-End_Error = 100 100 99 near_thresh 0 High_Fly_Writes = 1 1 0 near_thresh 132 Airflow_Temperature_Cel = 59 62 45 near_thresh 41 Temperature_Celsius = 41 38 0 ok 41 Hardware_ECC_Recovered = 58 43 0 ok 174433234 No SMART attributes are FAILING_NOW ============================================================================ ** Changed attributes in files: /tmp/smart_start_sdb /tmp/smart_finish_sdb ATTRIBUTE NEW_VAL OLD_VAL FAILURE_THRESHOLD STATUS RAW_VALUE Raw_Read_Error_Rate = 118 113 6 ok 177462074 Spin_Retry_Count = 100 100 97 near_thresh 0 End-to-End_Error = 100 100 99 near_thresh 0 High_Fly_Writes = 1 1 0 near_thresh 173 Airflow_Temperature_Cel = 59 61 45 near_thresh 41 Temperature_Celsius = 41 39 0 ok 41 No SMART attributes are FAILING_NOW ============================================================================ ** Changed attributes in files: /tmp/smart_start_sdb /tmp/smart_finish_sdb ATTRIBUTE NEW_VAL OLD_VAL FAILURE_THRESHOLD STATUS RAW_VALUE Raw_Read_Error_Rate = 114 118 6 ok 58789383 Seek_Error_Rate = 65 64 30 ok 3153602 Spin_Retry_Count = 100 100 97 near_thresh 0 End-to-End_Error = 100 100 99 near_thresh 0 High_Fly_Writes = 1 1 0 near_thresh 189 Airflow_Temperature_Cel = 59 61 45 near_thresh 41 Temperature_Celsius = 41 39 0 ok 41 Hardware_ECC_Recovered = 60 59 0 ok 58789383 No SMART attributes are FAILING_NOW What about that "Hardware_ECC_Recovered" increases coming and going. Doesn't sound good, but is it just me being the worrying kind? And what about the near_thresh of spin-retry and others. Need I worry? /Lars Olof

Thanks for letting me know, Joe. /Lars Olof