January 7, 201016 yr I noticed today that one of my drives has been disabled by unraid and has DISK_DSBL beside it in the unraid menu. I've read a few posts/wiki articles suggesting I run a smartctl report on the drive but I can't because the drive is disabled: Smartctl open device: /dev/sdc failed: No such file or directory root@Jewnas:/dev# ls sd* sda sda1 sdb sdb1 sdd sdd1 sde sde1 sdf sdf1 sdg sdg1 sdh sdh1 sdc is the drive in question. I'm also having a hard time getting a useful syslog, I haven't rebooted in about 2 months so it's 19 megs (I get hundreds of hack attempts a day). Is there a way to parse the log to eliminate failed ssh login attempts? I think that the drive is empty anyway and it has been taken out of the array (I have lots of free space and this was one of the later drives to b added when I built this a few months ago). See below for an unMenu printout of the drives: St Disk ? Dev Type ID Share Reads Writes Err Size % Used Free Usage Notes Temp Spin Info OK parity sdg WD 1.5T 3,439,475 3,433,575 0 1.5T 36C § sm hd sy sh OK disk1 sdf WD 1.5T 1,780,470 931 0 1.5T 97% 48G 32C — sm hd sy sh OK disk2 sdh Samsung 750G 650,855 514 0 750G 85% 114G * — sm hd sy sh OK disk3 sde WD 1T 6,843,305 2,830,679 0 1T 26% 745G 31C — sm hd sy sh DSBL disk4 sdc Hitachi 500G 6,843,305 2,830,679 2 500G Empty 500G * sm hd sy sh OK disk5 sdd WD 1T 2,419,343 9,156,446 0 1T 16% 843G 31C sm hd sy sh Total 4.75T 52.6% 2.25T Up Dn UN -- sda Unknown 1,702 0 * sm hd sy sh BT flash sdb Unknown 4G 2,003 1,498 4G 3% 4G sm hd sy sh No big deal if I have to replace the drive, I have a 1tb drive I just bought on the way home but before installing it and removing the drive with the error I'd like to make sure this is the correct/necessary course of action. Thoughts appreciated...
January 7, 201016 yr I noticed today that one of my drives has been disabled by unraid and has DISK_DSBL beside it in the unraid menu. I've read a few posts/wiki articles suggesting I run a smartctl report on the drive but I can't because the drive is disabled: Smartctl open device: /dev/sdc failed: No such file or directory root@Jewnas:/dev# ls sd* sda sda1 sdb sdb1 sdd sdd1 sde sde1 sdf sdf1 sdg sdg1 sdh sdh1 sdc is the drive in question. I'm also having a hard time getting a useful syslog, I haven't rebooted in about 2 months so it's 19 megs (I get hundreds of hack attempts a day). Is there a way to parse the log to eliminate failed ssh login attempts? I think that the drive is empty anyway and it has been taken out of the array (I have lots of free space and this was one of the later drives to b added when I built this a few months ago). See below for an unMenu printout of the drives: St Disk ? Dev Type ID Share Reads Writes Err Size % Used Free Usage Notes Temp Spin Info OK parity sdg WD 1.5T 3,439,475 3,433,575 0 1.5T 36C § sm hd sy sh OK disk1 sdf WD 1.5T 1,780,470 931 0 1.5T 97% 48G 32C — sm hd sy sh OK disk2 sdh Samsung 750G 650,855 514 0 750G 85% 114G * — sm hd sy sh OK disk3 sde WD 1T 6,843,305 2,830,679 0 1T 26% 745G 31C — sm hd sy sh DSBL disk4 sdc Hitachi 500G 6,843,305 2,830,679 2 500G Empty 500G * sm hd sy sh OK disk5 sdd WD 1T 2,419,343 9,156,446 0 1T 16% 843G 31C sm hd sy sh Total 4.75T 52.6% 2.25T Up Dn UN -- sda Unknown 1,702 0 * sm hd sy sh BT flash sdb Unknown 4G 2,003 1,498 4G 3% 4G sm hd sy sh No big deal if I have to replace the drive, I have a 1tb drive I just bought on the way home but before installing it and removing the drive with the error I'd like to make sure this is the correct/necessary course of action. Thoughts appreciated... If the drive is un-responsive, all you can do is stop the array, power down, verify a connector to it did not come loose. If there is any chance that is all it was, power back up see if you can run a smart report. If still un-responsive, power down once more, replace it, power up, press "Start" to rebuild your data onto the new drive. Whatever you do, DO NOT press the button labeled "restore" as it has nothing to do with restoring data. (there is an exception, but all your disks must be working) If it was just a loose cable on the old drive, you can use the trust my parity procedure as described in the wiki to re-enable the old drive. DO NOT USE IT (the "trust" procedure) IF YOU REPLACE A DRIVE OR IF YOU HAVE A FAILED DRIVE or YOU WILL LOSE THE CONTENTS OF THE DRIVE. If you replace the drive, use the "Start" button to begin the process of rebuilding onto the new drive.
February 5, 201016 yr Same problem, an empty 500 GB disabled "DISK_DSBL". Restart the array, still remain the same. It seems to spin however. Doesn't know how to run a smartctl. Why should I search for smartctl after another search? Have I to search for something else? People here doesn't know how to link something? I want to know : - how may I this disk un-disabled. - how to remove this disk (it was empty, it's still empty, but I want to remove it without replacement, good idea, isn't it?) BTW, is it a clean process to stop unraid? I've put a cache_dirs in the "go" file, and after this I've just to wait 65minutes if I stop the unraid just after a start. It's terrible. Even if I push the power on button on the tower, it's the same. And worse, even if I launch a "cache_dirs -q". Does it seems a weird idea, a stop button that stop it (by the web or on the tower)? If it was just a loose cable on the old drive, you can use the trust my parity procedure as described in the wiki to re-enable the old drive. No link? I have to search for it, but I already made a search (what a painful process) to be here! DO NOT USE IT (the "trust" procedure) IF YOU REPLACE A DRIVE OR IF YOU HAVE A FAILED DRIVE or YOU WILL LOSE THE CONTENTS OF THE DRIVE. What will that mean? Sometime I will lose the contents, sometime not?
February 5, 201016 yr Same problem, an empty 500 GB disabled "DISK_DSBL". Restart the array, still remain the same. It seems to spin however. Spinning does not matter if the data cannot be read. Doesn't know how to run a smartctl. Why should I search for smartctl after another search? Have I to search for something else? People here doesn't know how to link something? I'll save my return rant for later... Right now, to more completely help you, you can attach a syslog. Since you are using unMENU, go to the syslog page on it, click on the link to download it, then attach it to your next post. Since you are using unMENU, you have the ability to run smartctl reports on the disk management page on it. The button is probably labeled "Smart Status" or something like that depending on the version of unMENU you are using. You also have the ability on it to run"hdparm" commands. That button will let you know if the server can communicate with the disk at all. You don't have to search anywhere... If you do not have the current plug-in for unMENU, you can type smartctl -a -d ata /dev/sdc I want to know : - how may I this disk un-disabled. Disks are disabled any time any write to them fails. For your disk to be disabled, an attempt to write to it was made and that write failed. - how to remove this disk (it was empty, it's still empty, but I want to remove it without replacement, good idea, isn't it?) Yes, if you are not going to be replacing it, you can remove it if it is empty. To do that you will need to set a new disk configuration and completely re-compute parity without the drive you are removing. (Instructions will follow) BTW, is it a clean process to stop unraid?Yes, press the "Stop" button on the web-management interface. Since you have unMENU, it too has a "Stop" button on its "Array Management" page. I've put a cache_dirs in the "go" file, and after this I've just to wait 65minutes if I stop the unraid just after a start. It's terrible. If using cache_dirs, or ANY OTHER ADD-ON THAT USES THE DISKS, you MUST stop them from accessing the disks before stopping the array. It is doing its best to PROTECT YOUR DATA by NOT stopping in the middle of your use of it. Even if I push the power on button on the tower, it's the same.If you are unable to stop the server when pressing the power button on the server, then something is very unusual. We can't tell what is happening, since you did not attach a syslog.... you did not even tell us what version of unRAID you are running. Did you try holding the power button depressed for 10 seconds or so... Most power supplies will shut down if you do that. And worse, even if I launch a "cache_dirs -q".What happens when you type that command? It will kill any running cache_dirs program. But, we don't know if it is still running or not. What happens when you change directory to the folder where you put cache_dirs and then type cache_dirs -q ? ? ? Does it seems a weird idea, a stop button that stop it (by the web or on the tower)?They are there... and they will stop it. If running cache_dirs and using unRAID you will need to press "Stop" twice on the unRAID web-management interface, the second time must be a few seconds after the first. If you are running unMENU there is a "Stop" button on its array management page. It too will "Stop" the array. If by some chance you are actually referring to powering down the array, and not stopping the array (two entirely different operations) then the ability to power down from the unRAID interface depends on your BIOS. You must first "Stop" the array and then use the "Powerdown" button that will then appear. On some hardware you will be able to stop the array, and then shut-down linux, but not power off the power supply. The trust-my-parity process is NOT one you can use with a failed drive. Since you wish to remove the failed drive you must 1. Post a syslog. do this before you power down the array. This is critical for anybody to assist you. 2. power down the array (pull the power plug from the wall if you can't do it any other way, butt odds are good if you hold the power button down a few seconds it will ) 3. un-plug the failed drive physically 4. Power up the array. it will still show the array as DISK_DSBL. Go to the devices page and un-assign the failed disk. 5. Go back to the main unRAID web-interface. Check the checkbox under the button labeled as "restore" It is actually a "Set Disk Configuration" button. It will save a new disk configuration with only the remaining assigned and working drives. It will immediately invalidate any prior parity calculations. 6. Press the "Start" button. The array will start and will begin a full parity calculation based on the new disk configuration you set by pressing the button labeled "restore" (Again, it is very poorly named, it is a "Set Disk Configuration" button.) When parity calcs are complete, you will be back as you were previously, but before the disk failed and subsequently removed physically by you. You will be without parity protection until this new parity calculation is complete. Now that I've helped you... my "rants" in return: You did not tell us what version of unRAID you are running. (Makes a HUGE difference in how it reacts to stopping it while it is busy) You did not post a syslog, and if you did ANY search of any kind, for almost anybody having a disabled disk, it is the first item requested. You did not post the errors you got when trying commands, only that they did not work. We have NO idea what exactly you typed. Oops.. you did in your first post... it is /dev/sdc. You did not give me time to get a full 8 hours of sleep (you must be in a different time-zone then me, or need less shut-eye ) You did not go to the wiki (link in my signature) and search there for "trust" as it would point you to the exact "trust-my-parity" procedure I was referring to. (It does not apply to your situation, so do not use it) you complained about "people" here not providing a link when referencing a process of procedure, yet I was the only person who had responded... Therefore, you were complaining to me about me. I provide thousands of links.... just do a search... but it makes no sense to link anywhere until we know the actual issue. If you have not yet found it, look in the wiki: http://lime-technology.com/wiki/index.php?title=Unofficial_Documentation specifically here: http://lime-technology.com/wiki/index.php?title=Troubleshooting So... grab a syslog, post it, it will tell what is happening, then power down as instructed. It does not sound as if you've tried re-seating the connectors on the failed drive, since you are still powered up. Joe L.
Archived
This topic is now archived and is closed to further replies.