July 23, 201114 yr Hello, I've just checked my System only to discover the following: PARITY NOT VALID: DISK_DSBL label in both unMenu & Main Lime-Tech. I lost a PATA disk last week which was successfully recovered with a "new" SATA Disk (Thank You) and I then ran a complete Parity Check via Command Line (per Joe L.) with No Change to check for any errors (none found). Parity was Valid as of 7/19. System includes: unRaid Version 4.7 - PRO License Foxconn P41A-G LGA 775 Intel G41 ATX Intel Motherboard Intel Core2 6600 CPU WINTEC AMPO 2GB (2 x 1GB) DDR2 800 (PC2 6400) 3 SATA Drives (1.5TB Parity - ST31500341AS) & 2x1TB Drives) 5 PATA Drives (2x250; 1x300; & 2x500's - Mixed Maxtor/WDC/Seagate) Syslog Attached for Review. Please help, I had intended to upgrade the Parity Drive to 2TB and recycle the current Parity Drive, but a failure w/o warning is disconcerting! Please advise on the most approprate steps asap! Thanks, Dave syslog-2011-07-22.zip
July 23, 201114 yr Author Anyone have a suggestion, I'm getting a bit desparate Dave Update - Up to ~ 5 minutes ago (11:15 MDT) I could access my drives via the Network, now they are failing direct access \\10.0.0.50\disk1...disk8, flash, Movies etc shares). unMenu still reporting DSK_DSBL for Parity Drive as does Lime-Tech Main, but with drives "spinning" I'm not getting any Temps (0 degrees C) for Parity or Disk1 whereas Disk2/3/etc. show nominal numbers (26-32 degrees C). This system with exception of 1st Drive failure last week has been super stable for a number of years. I did upgrade to the Foxconn Board & Intel Core2 last fall but more for the additional ports (IDE/SATA) and d/t "flaky" onboard network card. Syslog is continuing to show Input/Output errors w/exception of spindown. See Syslog attached to original post. Do I need to shut down completely, restart, recheck Parity, or ?
July 23, 201114 yr disk0 and disk1 show errors in the syslog. Do they share a power splitter or PCI-SATA card?
July 24, 201114 yr Author disk0 and disk1 show errors in the syslog. Do they share a power splitter or PCI-SATA card? Thanks, I've checked and it does not appear so, but with the 3rd SATA drive now installed, I just might be running out of power. I had hoped to move my data off of the PATA drives (probably 2-5 years old and probably at/near/beyond MTBF) since the 1st one failed last week. Power Supply is 430 Watts XP550 Thermaltake. Dave
July 24, 201114 yr Via PM I learned you are using the round IDE cables. They are known to cause all sorts of problems. Since PATA has become uncommon this is not discussed much anymore, but the troubleshooting wiki page included warnings to avoid them. You definitely need to go back to the ornery flat ribbon cables, or better yet, go all SATA.
July 24, 201114 yr Look here (I wrote this a long time ago): http://lime-technology.com/wiki/index.php?title=Troubleshooting#Hard_drive_failures and here (I talk about the round cables in the 5th paragraph) http://lime-technology.com/forum/index.php?topic=1483.msg9973#msg9973
July 24, 201114 yr Author Via PM I learned you are using the round IDE cables. They are known to cause all sorts of problems. Since PATA has become uncommon this is not discussed much anymore, but the troubleshooting wiki page included warnings to avoid them. You definitely need to go back to the ornery flat ribbon cables, or better yet, go all SATA. Look here (I wrote this a long time ago): http://lime-technology.com/wiki/index.php?title=Troubleshooting#Hard_drive_failures and here (I talk about the round cables in the 5th paragraph) http://lime-technology.com/forum/index.php?topic=1483.msg9973#msg9973 Thanks again Brian, I will change the cables out, though as you noted, it is probably better to go all SATA. I guess my remaining questions are: 1 - Do I attempt to restart the current array (perhaps change the power cabling - Parity (sda) and Disk1 (sdb) are adjacent to each other) and hopefully get Parity restored, then perhaps re-created. Then attempt to copy the data off of several of my smaller PATA Drives first, take them out and attempt to get a "new" Parity. OR 2 - Replace the current Parity Drive immediatedly with the WD20EARS drive on the off chance that the current Parity (1.5 TB Seagate) has actually gone BAD. OR 3 - Some other process At the moment, Parity & Drive 1 are showing Red (Parity) Green (sdb) and according to Lime-Tech Main "spinning" but w/"0" degrees C showing (how if actually spinning ?). I just attempted to access disk1 (\\10.0.0.50\disk1) and saw the various shares in the top level menu, but as soon as I attempted to view Movies, it locked up - not responding. Tried again and able to see Minidlna (1 file in sub-directory) and a couple of others, but not movies. Can this info be "cached" somewhere I do have some "flat" PATA cables, so will try to use them, though I think they may cause higher temps d/t airflow issues. Again Thanks - if anyone has a suggestion of how I should proceed, please let me know. I don't really like being "naked" w/o Parity especially if the drives may be getting flaky! Dave Edit: I just looked at my Lime Technology Main Page and see some write errors on the Parity Drive. I'm not sure if that is significant, but I don't recall seeing any errors before. I've attached a picture of this for your information.
July 24, 201114 yr The first thing I would do is run a smart report on the parity disk. I think that the errors you are seeing are somehow related to the problems you're having with the other disks. But if you are seeing reallocated sectors, pending sectors, or failed overall health check, that is another matter. You can use myMain or run it from the command line. See the troubleshooting link in my sig for instructions. After ruling out a serious disk problem with the parity disk, I'd suggest stopping the array and making a backup copy of the "config" folder on your flash disk. Then run "initconfig" command from a telnet prompt. Shutdown the machine. Fix your pata cables. Double check connections to all of your SATA drives too (power and SATA). Boot back up. Verify that all are recognized in the BIOS before allowing unRAID to boot fully. Once unRAID boots, depending on version, the drives may or may not be assigned to the array slots. Assign all of the data disks BUT NOT PARITY. Leave the parity slot empty. (It is easy to screw up and put disk1 in the parity slot, don't do it!) Start the array (with NO parity disk). Verify the contents of each and every disk (disk1 - disk7). AFTER you have confirmed you can read every disk, copy 10G+ to each disk, and then back to your workstation. (If you can't read from a disk, you must not write to ANY disk). Once you are confident that the disks are all working and your data is in place, stop the array, assign parity, and start the array. Your system will rebuild the parity. This is a pretty good burn in for all of your disks. There are ways I could tell you to shortcut the parity build, but I'd like to see parity build successfully on your system. Then you should be good. Report back with any questions or if anything seems to go wrong.
July 24, 201114 yr Author The first thing I would do is run a smart report on the parity disk. I think that the errors you are seeing are somehow related to the problems you're having with the other disks. But if you are seeing reallocated sectors, pending sectors, or failed overall health check, that is another matter. You can use myMain or run it from the command line. See the troubleshooting link in my sig for instructions. Hi Brian, Sorry didn't see your post last night, tried the above utilizing both myMain (unMenu) and your command line from the Troubleshooting guide. I probably don't quite know what to look for in myMain but this is what came back from the command line. "A mandatory SMART command failed: exiting. To continue, add one or more '-T permissive' options." I'm not exactly sure how I should proceed at this point. Dave
July 24, 201114 yr The first thing I would do is run a smart report on the parity disk. I think that the errors you are seeing are somehow related to the problems you're having with the other disks. But if you are seeing reallocated sectors, pending sectors, or failed overall health check, that is another matter. You can use myMain or run it from the command line. See the troubleshooting link in my sig for instructions. Hi Brian, Sorry didn't see your post last night, tried the above utilizing both myMain (unMenu) and your command line from the Troubleshooting guide. I probably don't quite know what to look for in myMain but this is what came back from the command line. "A mandatory SMART command failed: exiting. To continue, add one or more '-T permissive' options." I'm not exactly sure how I should proceed at this point. Dave Please post exactly the command you entered and exactly the smartctl output.
July 24, 201114 yr Author Please post exactly the command you entered and exactly the smartctl output. Hi Brian, from your Troubleshooting Guide, I did a telnet to 10.0.0.50 and a login to Tower - Root root@tower:~# smartctl -a -d ata /dev/sda | todos >/boot/smart.txt Response: Smartctl: Device Read Identity Failed (not an ATA/ATAPI device) A mandatory SMART command failed: exiting. To continue, add one or more '-T permissive' options. I've also looked at myMain and in the "sm" key I have a similar response: smartctl -a -d ata /dev/sda (parity) smartctl 5.39.1 2010-01-28 r3054 [i486-slackware-linux-gnu] (local build)Copyright © 2002-10 by Bruce Allen, http://smartmontools.sourceforge.netSmartctl: Device Read Identity Failed (not an ATA/ATAPI device)A mandatory SMART command failed: exiting. To continue, add one or more '-T permissive' options. The remainder of the myMain keys are essentially blank. I've tried to attach a zip copy (but its too big) for your or others review. I may have to do a modify and/or separate post. Dave Edit: - Sorry all the versions of myMain screen shot seem to be too big to post even zipped.
July 25, 201114 yr Please post exactly the command you entered and exactly the smartctl output. Hi Brian, from your Troubleshooting Guide, I did a telnet to 10.0.0.50 and a login to Tower - Root root@tower:~# smartctl -a -d ata /dev/sda | todos >/boot/smart.txt Response: Smartctl: Device Read Identity Failed (not an ATA/ATAPI device) A mandatory SMART command failed: exiting. To continue, add one or more '-T permissive' options. I've also looked at myMain and in the "sm" key I have a similar response: smartctl -a -d ata /dev/sda (parity) smartctl 5.39.1 2010-01-28 r3054 [i486-slackware-linux-gnu] (local build)Copyright © 2002-10 by Bruce Allen, http://smartmontools.sourceforge.netSmartctl: Device Read Identity Failed (not an ATA/ATAPI device)A mandatory SMART command failed: exiting. To continue, add one or more '-T permissive' options. The remainder of the myMain keys are essentially blank. I've tried to attach a zip copy (but its too big) for your or others review. I may have to do a modify and/or separate post. Dave Edit: - Sorry all the versions of myMain screen shot seem to be too big to post even zipped. Why do you think your disk is sda? The myMain screen should show you the device. But you could also just run it by clicking the "sm" hyperlink in the "Info" column. Or if you are sure that the drive is sda, try this command: smartctl -a -/dev/sda | todos >/boot/smart.txt
July 25, 201114 yr Author Why do you think your disk is sda? The myMain screen should show you the device. But you could also just run it by clicking the "sm" hyperlink in the "Info" column. Or if you are sure that the drive is sda, try this command: smartctl -a -/dev/sda | todos >/boot/smart.txt Hi Brian; sda is listed as Parity in the Devices list in my Lime-Tech Devices page and in the unMenu screen for myMain and the general base screen (//tower:8080/ or in my case //10.0.0.50:8080/). It was Parity when I rebuilt my disk2 (hda) that failed (worked) and it is also the largest disk in my system and if I recall must be Parity d/t that requirement. I have run your command and have the following report, plus the --all command after invoking a -test=short: smartctl 5.39.1 2010-01-28 r3054 [i486-slackware-linux-gnu] (local build) Copyright © 2002-10 by Bruce Allen, http://smartmontools.sourceforge.net Device: /1:0:0:0 Version: scsiModePageOffset: response length too short, resp_len=47 offset=50 bd_len=46 >> Terminate command early due to bad response to IEC mode page A mandatory SMART command failed: exiting. To continue, add one or more '-T permissive' options. Hope the helps clarify the problem. Dave
July 25, 201114 yr Why do you think your disk is sda? The myMain screen should show you the device. But you could also just run it by clicking the "sm" hyperlink in the "Info" column. Or if you are sure that the drive is sda, try this command: smartctl -a -/dev/sda | todos >/boot/smart.txt Hi Brian; sda is listed as Parity in the Devices list in my Lime-Tech Devices page and in the unMenu screen for myMain and the general base screen (//tower:8080/ or in my case //10.0.0.50:8080/). It was Parity when I rebuilt my disk2 (hda) that failed (worked) and it is also the largest disk in my system and if I recall must be Parity d/t that requirement. I have run your command and have the following report, plus the --all command after invoking a -test=short: smartctl 5.39.1 2010-01-28 r3054 [i486-slackware-linux-gnu] (local build) Copyright © 2002-10 by Bruce Allen, http://smartmontools.sourceforge.net Device: /1:0:0:0 Version: scsiModePageOffset: response length too short, resp_len=47 offset=50 bd_len=46 >> Terminate command early due to bad response to IEC mode page A mandatory SMART command failed: exiting. To continue, add one or more '-T permissive' options. Hope the helps clarify the problem. Dave There should be no dash before the /dev/sda. smartctl -a /dev/sda | todos >/boot/smart.txt You can try wtih the -T Permissive option. smartctl -a -T Permissive /dev/sda | todos >/boot/smart.txt What is the motherboard? What controller is this running off of?
July 25, 201114 yr Author There should be no dash before the /dev/sda. smartctl -a /dev/sda | todos >/boot/smart.txt You can try wtih the -T Permissive option. smartctl -a -T Permissive /dev/sda | todos >/boot/smart.txt What is the motherboard? What controller is this running off of? Hi Brian, unRaid Version 4.7 - PRO License Foxconn P41A-G LGA 775 Intel G41 ATX Intel Motherboard Intel Core2 6600 CPU WINTEC AMPO 2GB (2 x 1GB) DDR2 800 (PC2 6400) 3 SATA Drives (1.5TB Parity - ST31500341AS) & 2x1TB Drives) - edit - all from main board no add-in SATA 5 PATA Drives (2x250; 1x300; & 2x500's - Mixed Maxtor/WDC/Seagate) - edit - 1 (hdb) from main IDE 3 from Promise card the "-" came from it reporting a need to try -h and from the troubleshooting guide, sorry. running: smartctl -a -T Permissive /dev/sda | todos >/boot/smart.txt yields smartctl 5.39.1 2010-01-28 r3054 [i486-slackware-linux-gnu] (local build) Copyright © 2002-10 by Bruce Allen, http://smartmontools.sourceforge.net =======> INVALID ARGUMENT TO -T: Permissive =======> VALID ARGUMENTS ARE: normal, conservative, permissive, verypermissive <======= Use smartctl -h to get a usage summary for fun I tried the smartctl -a -T verypermissive /dev/sda | todos >/boot/smart.txt that it suggested and got smartctl 5.39.1 2010-01-28 r3054 [i486-slackware-linux-gnu] (local build) Copyright © 2002-10 by Bruce Allen, http://smartmontools.sourceforge.net Device: /1:0:0:0 Version: scsiModePageOffset: response length too short, resp_len=47 offset=50 bd_len=46 >> Terminate command early due to bad response to IEC mode page Log Sense failed, IE page [scsi response fails sanity test] Error Counter logging not supported scsiModePageOffset: response length too short, resp_len=47 offset=50 bd_len=46 Device does not support Self Test logging Dave
July 25, 201114 yr Author Not to be "quip" but the response that it "fails sanity test" seems truly appropriate
July 25, 201114 yr The first thing I would do is run a smart report on the parity disk. I think that the errors you are seeing are somehow related to the problems you're having with the other disks. But if you are seeing reallocated sectors, pending sectors, or failed overall health check, that is another matter. You can use myMain or run it from the command line. See the troubleshooting link in my sig for instructions. After ruling out a serious disk problem with the parity disk, I'd suggest stopping the array and making a backup copy of the "config" folder on your flash disk. Then run "initconfig" command from a telnet prompt. Shutdown the machine. Fix your pata cables. Double check connections to all of your SATA drives too (power and SATA). Boot back up. Verify that all are recognized in the BIOS before allowing unRAID to boot fully. Once unRAID boots, depending on version, the drives may or may not be assigned to the array slots. Assign all of the data disks BUT NOT PARITY. Leave the parity slot empty. (It is easy to screw up and put disk1 in the parity slot, don't do it!) Start the array (with NO parity disk). Verify the contents of each and every disk (disk1 - disk7). AFTER you have confirmed you can read every disk, copy 10G+ to each disk, and then back to your workstation. (If you can't read from a disk, you must not write to ANY disk). Once you are confident that the disks are all working and your data is in place, stop the array, assign parity, and start the array. Your system will rebuild the parity. This is a pretty good burn in for all of your disks. There are ways I could tell you to shortcut the parity build, but I'd like to see parity build successfully on your system. Then you should be good. Report back with any questions or if anything seems to go wrong. Did you do all of this? Please do and post a full syslog again after a fresh boot.
July 25, 201114 yr Author Report back with any questions or if anything seems to go wrong. Did you do all of this? Please do and post a full syslog again after a fresh boot. Sorry, when I failed to get a "smart" report, I stopped & "reported back". It seems that I need to just continue past the Parity Smart report and see what happens. Hopefully I can get that done this evening and give you a new Syslog to work with. Thanks again for your patience and help, this has been quite illucidating! Dave
July 25, 201114 yr Author Report back with any questions or if anything seems to go wrong. Did you do all of this? Please do and post a full syslog again after a fresh boot. Hi Brian, Quick update, Server Stopped, All Cables changed, Power checked (dgaschk was correct the power leads were linked upstream, so I've changed them to another lead), re-set all the disks to their prior location (when it restarted, they were in a jumbled order), Parity is unassigned, all disks were readable, ~ 12GB is currently being copied (that will take a bit unless someone has a quicker method). Again thanks for the feedback. I hope to have something more definitive to report, though suspect at this rate it will be a couple of days. Please check for a PM indicating a status update. Dave
July 25, 201114 yr Report back with any questions or if anything seems to go wrong. Did you do all of this? Please do and post a full syslog again after a fresh boot. Hi Brian, Quick update, Server Stopped, All Cables changed, Power checked (dgaschk was correct the power leads were linked upstream, so I've changed them to another lead), re-set all the disks to their prior location (when it restarted, they were in a jumbled order), Parity is unassigned, all disks were readable, ~ 12GB is currently being copied (that will take a bit unless someone has a quicker method). Again thanks for the feedback. I hope to have something more definitive to report, though suspect at this rate it will be a couple of days. Please check for a PM indicating a status update. Dave I have no explanation as to why you couldn't run the smart report. Maybe was related to the cabling problems. But after the reboot this is looking positive. Hopefully all disks will check out, you can add parity, and it will rebuild successfully. Fingers crossed!
July 25, 201114 yr Author I have no explanation as to why you couldn't run the smart report. Maybe was related to the cabling problems. But after the reboot this is looking positive. Hopefully all disks will check out, you can add parity, and it will rebuild successfully. Fingers crossed! Hi Brian - "DITTO" Quick update, overnight able to copy ~ 12GB to 4 of the 7 drives, the 3 smaller PATA's don't have that much space, so I've sent several smaller files over to them this morning. Am now copying them back to my XP drive to test Read/Write. No Errors thus far ! Since I want to elimate the PATA Drives (for reasons stated above and probably for energy savings (less even in standby mode)) what is the best way to copy all of that data to my SATA's? If there are any duplicates (movies/videos/*.db files/???) is there a way to combine while copying and not have a halt or ? by the process (ie Windows)? I seem to recall a DOS command but am not sure of a Linux command. Dave Edit - Read/Write - Read Back complete w/o errors. Parity Sync underway for ~ 3 hours w/estimated 1700 minutes left (~ 9.4% done). No errors thus far - here's hoping! - D
July 26, 201114 yr I have no explanation as to why you couldn't run the smart report. Maybe was related to the cabling problems. But after the reboot this is looking positive. Hopefully all disks will check out, you can add parity, and it will rebuild successfully. Fingers crossed! Hi Brian - "DITTO" Quick update, overnight able to copy ~ 12GB to 4 of the 7 drives, the 3 smaller PATA's don't have that much space, so I've sent several smaller files over to them this morning. Am now copying them back to my XP drive to test Read/Write. No Errors thus far ! Since I want to elimate the PATA Drives (for reasons stated above and probably for energy savings (less even in standby mode)) what is the best way to copy all of that data to my SATA's? If there are any duplicates (movies/videos/*.db files/???) is there a way to combine while copying and not have a halt or ? by the process (ie Windows)? I seem to recall a DOS command but am not sure of a Linux command. Dave Edit - Read/Write - Read Back complete w/o errors. Parity Sync underway for ~ 3 hours w/estimated 1700 minutes left (~ 9.4% done). No errors thus far - here's hoping! - D Hope it completes for you! Those nasty round IDE cables are killers. Make sure to run a parity check after the parity build. It will be your final proof that the array is protected.
July 26, 201114 yr Author Hope it completes for you! Those nasty round IDE cables are killers. Make sure to run a parity check after the parity build. It will be your final proof that the array is protected. Hi Brian, me TOO I just looked and its only about 21% complete, so despite its projection from 1800 minutes @ 1245, its got a long way to go - probably all night and even maybe tomorrow too. I do want to follow with a Parity Check, but if all goes well, have thoughts of changing to the larger 2TB WD20EARS for parity (if it fails to create a Parity, I can fall back to the current one (as I would un-assign it as Parity prior to putting the new one in place)). Does that make sense or should I have the current Parity checked first? It would give me enough space to eliminate all of the PATA Drives (remove 5 for 1, good trade), because I would be able to add the current Parity (1.5TB) drive to the Array (after the 2TB drive is in place w/good Parity). Did you see my ? about copying my PATA data to my SATA drive(s)? I do think that I have some file duplicates to eliminate (Movies Share). Thanks again for your help! Dave
July 26, 201114 yr Your parity build speed should go up considerably once you get past the smaller PATA drives. Hopefully you will be there soon.
July 26, 201114 yr Author Your parity build speed should go up considerably once you get past the smaller PATA drives. Hopefully you will be there soon. Hope so, 13000KB/sec w/1400 minutes left & 3 PATA Drives to still clear. Seems really slow, but probably pretty normal. I think all of the drives are working pretty well, though speed is not their forte! Have a good night, will update in AM. Dave Edit - PARITY VALID And Wonder of Wonders I have a SMART Report (attached)! I've also enclosed a copy of my Syslog. Brian if you have a chance, please evaluate. Thank you again so much for your help and patience! Warmly, Dave smart.txt smart.txt syslog_2.txt
Archived
This topic is now archived and is closed to further replies.