March 1, 200818 yr Ok so I've filled my newly built Unraid server. I've got a 1TB parity, 4x 500gb and 1x 750gb drives. So I read the manual for "upgrading a drive" and did the following. I shut down the server, unplugged one of the 500gb drives and plugged in a brand spanking new 2nd 1tb drive. I then rebooted. Upon logging on the server it said it was "upgrading disc". I then pressed start and it began (I think) to rebuild the data off the parity drive from the 500gb drive onto the new 1tb drive. This finished (I think) and rebooted on its own. When I logged back on the new 1TB drive was listed showing ~500 gb free (makes sense b/c that's about how much info would have been copied onto the new drive). Only thing is, the drive is red - not green? Its got a star next to it where the temp normally is listed. So then I click "start all drives". Then the new 1TB drive is suddenly listed as being a 2.1 TB drive?? huh?? (which unfortunately it is not) and it said the parity drive is not the largest drive. So then I reboot. Then it starts again and once again the new 1TB drive is listed as red and back to being the correct size. Well I check the drive from my workstation and I am able to read all the data on the drive and am even successful in writing to the drive (ie everything seems to be fine except the dot next to the drive is red - not green). I reboot x 2 and no change - the 1 TB drive is still red. So I think maybe I just need to do a parity sync. So I do the "system restore" now all the drives are green (except the parity drive) and I click start parity sync. Then it freezes or something and I reboot again. Now it states that the 1 TB drive is not formatted and needs to be formated!! (Please let the pain stop). So I throw in the towel and plug the original 500gb drive back in and unplug the 1 TB drive and reboot. The system won't start because it sees the "old" 500gb drive as a new drive and I"m forced to resync the parity ARRGGH so here I am -and my blood pressure is slowly coming back to normal and I'm sure my current parity sync will work and I'll have all my original data - But now I'm just back where I started - no more room on the array and an extra 1TB drive just begging to be included! Where did I go wrong? Was my parity likely not correct when I started the upgrade or something? I"m getting a 8 port sata PCI card that people here have recommended, but it won't be here for a week, and I need the space on my server now. Thanks in advance for your help
March 2, 200818 yr This finished (I think) and rebooted on its own. Reboot is not normal during a drive upgrade, there’s something wrong with your hardware, I’m sorry I can’t help more, maybe someone else has an idea.
March 2, 200818 yr First... NEVER EVER use the "Restore" button when an incomplete set of data drives is present and assigned. It does not "restore" anything. Instead, it "stores" a brand new configuration to the system.dat file based on the currently assigned drives on the devices page and then re-computes parity based on those, and ONLY those same drives. Pressing the "Restore" button forgets any previously calculated parity information by design. It erases any knowledge of any data from a failed or missing drive as if you said "Yes, a drive has failed, but I do not need any data that was on it, so re-calculate parity based on the remaining assigned drives so I can recover if another drive fails." So, unless you are removing a drive from an installation, and want to re-calculate parity so you could recover from any other failure, do not use the "Restore" button. When replacing one drive with another all you should have to do is start the array. It will do the rest. If your array "re-booted" then something went very wrong. It will not re-boot, ever, under any of its administrative functions, unless you press the "Reboot" button, and then it is only doing that because you asked it to, not because it needs to. (This is Linux, not Microsoft... Linux machines stay on-line for months and years. Re-boots are only to add/remove hardware) So, assuming your blood pressure is back to normal, and the old 500G drive is in place, and parity re-calculated, you can go through the steps once more. If your new drive is mis-reporting its size, or doing so after it heats up, that could be the cause of your crash and re-boot. It could also be power supply related, as a parity sync has all drives spinning. It could just as easily be a loose cable. Check those are still seated. I would go through these steps. 1. Stop the array. 2. Un-assign the drive you are going to replace. 3. Start the array... You will see it marked as missing, and the indicator is RED ... however you can still read and write to it. The fact that you can read and write does not mean all is OK, it means the RAID software is doing its job... protecting you when a disk has failed (or is missing) The RED indicator was telling you the truth the first time you attempted to upgrade. The new drive was NOT YET online. It was being loaded/re-built from parity and the other drives. 4. Stop the array once more. 5. Power down and substitute the new 1T drive for the old. 6. Power up. unRaid should notice the drive has been replaced and will allow you to start it (assuming the new 1T drive has the same or less blocks in size reported to unRaid as your current parity drive. If it is even one block bigger, it will suggest a parity-swap. The parity-swap feature is broken , a fix is available from Tom as version 4.2.2, and it will be fixed in the 4.3 release, due out soon) 7. Once powered up, the contents of the old 500Gig drive will be re-built on the new drive. While it is doing this, the RED indicator will be on. While it is doing this, you will be able to read and write to the new drive. It is actually writing to the parity drive as if the new drive was there and ready. This process of re-building the old data onto the new 1T drive will take many many hours, probably between 4 and 16, depending on the speed of your array. Do not interrupt it while it is doing this. (you can press the refresh button, but that is about it) If the process does not complete, or the array re-boots, the odds are high you have a hardware issue to deal with... At least on older releases, the "Refresh" button did nothing while it was clearing a new drive, and it appeared as if the web-interface was frozen. The array itself was not, it was just not responding to web-commands while it was dealing with the new drive. I don't know if this has changed recently, I'll have to try a disk replacement on my array to see how it reacts. (Have not replaced a drive in quite a while, and back then it was only to experiment and help find a bug in a very early version of unRaid) You do not need to press the "Check parity button" unless you simply want to exercise your disks, once a month or so to exercise them is probably plenty... other than that, unRaid calculates parity at all times unless you have no parity disk assigned. Joe L.
March 2, 200818 yr Very good advice above. Your brand new terabyte drive sounds defective. I am not a fan of this method of upgrading a drive, and your situation is a perfect example of why. There is no higher percentage of failure for any drive than in its first month, especially (I think) its first week. Good drive management dictates that any drive be thoroughly tested first, before being put online, in production. And this is especially true with unRAID, because as soon as any drive is pulled, the entire array is vulnerable. The last thing you want to do at that moment is add an unknown, untested drive to it. I personally believe in either testing the drive thoroughly before adding it to the array, or adding it and then running full tests. Only afterward would I trust it as a new data drive, or as an upgrade of an existing drive. There's one other possibility, and I would like to hear from anyone that has actually done this, expanded a 500GB drive to 1TB drive, to confirm that there were no problems. With most operating systems, expanding a partition can involve changing cluster size, expanding the FAT, or similar operations appropriate for a particular operating system. I don't know the mechanics of the Reiser file system, but it would not at all surprise me if there were some file system structure expansions necessary, when adding 500GB to it. This upgrade process just mirrors the original 500GB drive onto the terabyte drive, so the file system structures may not be able to address a terabyte of space. Can we hear from someone who has done this?
March 2, 200818 yr Author Thank you all so much for the replies. I've got another 2 hours of parity sync and I'll try it again. I definiitely don't think there is a problem with the new 1TB drive, I've used it for about a month as an external drive on my HTPC and its worked flawlessly. I'll let you know how it goes. Cheers
March 2, 200818 yr Double check the cabling! It sounds like the OS is getting some wrong drive information about the drive geometry - which may mean that the data cable is flakey. Try another one. Although unlikely, I woud not rule out a bad drive. Rebooting during the rebuild process is definitely not normal! A bad data cable might be at the heart of a dynamic reboot, but I would not think so. The only other thing I can think of is the power supply. Is it possible that the PSU is overworked? How big is your power supply? This raises an interesting question. If you are doing a drive replacement (due to upgrade or more to the point, for a failed disk), and the computer happens to reboot in the middle (power failure, for example), will unRAID pick up and resume the rebuild process? If not, what should a person do?
March 2, 200818 yr Author Ok so I've re-started the upgrade process. I followed the exact instructions as above. I'm now about 10 hrs into the process and here is what I see (screenshot) Please note that all drives are flashing green, excpt the new 1TB drive is still red. Is this normal behavior? Is it just not finished yet? The fact that the parity drive is spun down worries me. I would think it would need to be spinning during this whole process? 289 Error too, what's that about? Thanks Also,I checked all the cables. My PSU is a new Corsair 550watts which is clearly enough for only 6 drives. The only other think I can think of, is that my new 440 celeron with my outdated Asus bios. I get an "invalid CPU ucode press F1 to continue" message upon booting. If I press F1 everything seems to work fine. I realize I should update the bios, but not sure how do to do this in a uraid system.
March 2, 200818 yr The flashing icons indicate that the drives have spun down. When doing a rebuild NONE of the drive should be flashing! You should see the read counts incrementing an all the drives except one, and the write counts incrementing on the other one. Your I/O counts are tiny (each one represents a 4K block, so what you are showing there is nothing. It is not rebuilding, nor has it rebuilt, anything of any size. I will leave to others to explain the screen in more detail. I am not that experienced with these types of problems. It looks like to me that you had a bunch of errors and it gave up. Did you try replacing the data cable? Look at this thread http://lime-technology.com/forum/index.php?topic=1521.0 and find a zip file attached called smartctl.zip. In it is a tool that will show you a bunch of data the the drive keeps about itself that may provide some hints as to what is going on. Follow the directions to install and run it on your flakey drive. Post the results. I had that same CPU ID issue on another computer. I was able to create a DOS boot USB stick. It was pretty easy. There is an HP format tool (a quick Google search found this link http://files.extremeoverclocking.com/file.php?f=197). After you have your bootable USB, you can copy whatever files you need onto the USB stick to flash your BIOS. WARNING: DO NOT DO THIS ON YOUR UNRAID USB STICK - USE ANOTHER ONE! You should definitely update your BIOS!!
March 2, 200818 yr I just happened to have a spare 400 Gig drive I recently removed from one of my media players... To keep myself current with how unRaid reacts when you try to upgrade a disk I shut down my array, removed a 250 Gig drive that used to be in the "disk2" slot of my server and replaced it with the 400 Gig drive. First and most important, I went to the "Devices" page on the web-interface and did used the "print" feature of my browser to get a listing of the current disk assignments, before I made any changes. The reason is simple, sometimes Linux will scan hardware in a different order, and a disk that is /dev/hda today might end up as /dev/sda tomorrow if you replaced a IDE disk with an SATA disk. The same could also happen if you used a different cable, or connector, or disk controller. In my example here, since the controller and cable were the same, the linux device did not change. If it had changed, then it wold be critical to make sure the parity drive was still assigned the parity slot before continuing. When I powered up the array showed a "BLUE" indicator for disk2. In the command area it said: Stopped. Upgrading Disk. Both the "Start" and "Restore" buttons were greyed out, each had a checkbox below that would activate them. "Start" had as its description: Start will bring the array on-line, start Data-Rebuild, and then expand the file system. "Restore" had as its description: Restore will initialize the stored array configuration; all drives will appear as New, but data disk contents are not affected. edit: in newer versions of unRAID, the button labeled as "restore" was removed because it was poorly labeled. It is NOT used in the replacement of a new disk, so if it does not exist on your screen do not worry. You would not have used it. It does not "restore" anything... never did. Note: I did not have to use the device assignment page to assign the replacement drive. unRaid knew the drive was upgraded since the serial number was different. The linux devices did not change. If devices were re-named by Linux, odds are I would have a red indicator or two showing and it would be the first clue I need to verify the device assignment on the "Devices" page. The web-management page looked like this: If I were to mistakenly activate the "restore" button and press it, my array would forget any contents I used to have on disk2, probably would clear and format the new drive 2, and then bring the array on-line. Note... this is NOT what you want to do if you expect the old contents of disk2 to be restored to the new replacement disk2. As I have said in other posts, the "Restore" button is poorly labeled. It is really a "New configuration based on currently assigned drives" button. Do not use it when upgrading a disk or replacing a failed disk. If you do, you will lose your data, and if replacing a failed drive, no chance of getting it back. Since I do not want to lose the contents of disk2, and I do want to re-create the contents of my old disk2 to the new replacement drive, I want to use the "Start" button. This is exactly the same choice I would need to make if my disk2 had failed and I just replaced it with a new drive. Ok... I check the checkbox under start and press the "Start" button. The web-page refreshes itself once and looks like this: Notice, it just says "Starting" and shows that it is mounting all the disks. The only button available is the "Refresh" button. Unless I press it, I will not see any change on the screen.... I waited a few minutes to see if it would update itself... it did not. I logged in via telnet to see if the drives were indeed mounted. They were: Tower login: root Password: [Disconnect bypassed -- root login allowed.] Linux 2.6.22.5. root@Tower:~# mount /dev/sda1 on /boot type vfat (rw) none on /sys/fs/fuse/connections type fusectl (rw) /dev/md5 on /mnt/disk5 type reiserfs (rw,noatime,nodiratime) /dev/md8 on /mnt/disk8 type reiserfs (rw,noatime,nodiratime) /dev/md3 on /mnt/disk3 type reiserfs (rw,noatime,nodiratime) /dev/md4 on /mnt/disk4 type reiserfs (rw,noatime,nodiratime) /dev/md7 on /mnt/disk7 type reiserfs (rw,noatime,nodiratime) /dev/md6 on /mnt/disk6 type reiserfs (rw,noatime,nodiratime) /dev/md1 on /mnt/disk1 type reiserfs (rw,noatime,nodiratime) /dev/md2 on /mnt/disk2 type reiserfs (rw,noatime,nodiratime) shfs on /mnt/user type fuse (rw,nosuid,nodev) root@Tower:~# I pressed the "Refresh" button on the web-interface. It refreshed to show the array was on-line and the rebuild was in progress. It warned against stopping the Data-Rebuild as it would leave the array un-protected. (I'm guessing if you stopped it you could start it again from the beginning as long as all the remaining drives are working. Remember... we have one "missing/failed" drive... until we get a good parity drive calculation with the new drive, we will lose data if a second drive were to fail.) The web-interface looks like this... It shows my disk is already at 400 Gig, so it lied previously as it said it would expand the drive after restoring the contents. Instead, it expanded the drive first and is now restoring its contents. (Tom recently said that allows you to use the full size of the new drive even before its data is rebuilt.. Yup, while it is rebuilding my disk2 I can still get to it to read or write.) In about 377 minutes I should be able to press "Refresh" to see the drive rebuild complete. Joe L.
March 2, 200818 yr It is time for you to post a copy of your syslog. From what I can see from your screen shot, disk5 has failed. the syslog should give more clues. Log in via telnet and type: cp /var/log/syslog /boot/syslog.txt You can then browse to your flash drive to get a copy, or shut down and put the flash drive in a PC to attach a copy of syslog.txt to this thread. My disk2 rebuild has progressed some... as other have said, you should see the "write" count incrementing. Here is a new screen shot when I pressed "Refresh" once more: Only 348 minutes to go on my all IDE based array with the original Intel motherboard. Joe L.
March 2, 200818 yr Your screen shot seems to indicate the re-build process started, but did not get very far before the disk failed and the re-build process stopped. That explains why everything is "flashing" as all your drives have gone to sleep after being idle for a hour or so. (I'll bet they have been idle for a lot longer than that even, based on how little data was written to the replacement drive.) Joe L.
March 2, 200818 yr Joe - that is an EXCELLENT guide to how to upgrade a disk! I'm keeping that one "on speed dial" so that I can post a reference to others asking how to upgrade a disk. One very minor comment, I have noticed the screen with the "Starting" status (showing all the drives "mounting") will update and show all the drives mounted with their sizes shown. It will stay like that and you'll start to see some activity - writes to the parity drive and data drive. It can be in that state for well over a minute before going on to the next screen. I believe that what it is diong is formatting the disk and growing the file system. Interesting things happen to the reported freespace on the new drive during this time. I am still wondering (concerned) what would happen if there were a power outaqe while rebuilding? Would unRAID realize that there was a "rebuildus interruptus" and proceed to finish the job? Or let you begin the process again? The risk of a power failure during a 30 second operation like a BIOS update is small, but the chances during a drive rebuild that could take 12 hours is not so small.
March 2, 200818 yr Actually, I left the browser alone for quite a while when it first refreshed and said "Mounting" for all my disks. It did not refresh by itself after that to show the drives were mounted. I had to do the refresh myself. (I'm using firefox... perhaps it is different if using Internet-Explorer... but somehow, I doubt it) That is why I logged in via telnet to see the actual status... and then pressed the refresh button to see it on the web-interface too. If the re-build were to be interrupted, I think you would just have to start it again from the beginning. Just do NOT use the "Restore" button... (I think I said that already, but it needs repeating again, and again....) Joe L.
March 2, 200818 yr Thanks Joe! Yes, when I was doing it with IE I also had to keep refreshing the browser. rpf717 - how's it going?
March 2, 200818 yr Author Ok so I figured I'd try to to the rebuild process again. Stopped the array and powered down. Now the system freezes during the boot procees/POST or whatever its called. IT states Can't detect TPM device ! Initializing USB Controllers . . Done. USB Device(s): 1 Storage Device Auto-Detecting SATA 1. . . IDE Hard Disk Auto-Detecting SATA 5. . . IDE Hard Disk Auto-Detecting SATA 2 . . . It just stops there . . . If I reboot with the old 500gb drive in place, it boots up just fine. I've tried this with another SATA drive other than the "offending" 1TB drive and it too is not recongnized during the boot sequence - but it doesn't freeze. Unraid boots, but the disk is not recognized.
March 2, 200818 yr Author also, as far as the syslog is concerned. Where exactly am I typing that command line in. On my server it says Tower login, I then logged in with user name and password, and typed the command line as you mentioned, but the syslog doesn't appear in my flash drive. Am I doing this correctly? Sorry I've never used a Linux system before.
March 2, 200818 yr Recommendation #1. TAKE YOUR TIME. Recommendation #2. If you ask for help here, make sure you post the diagnostic materials requested. Recommendation #3. Make sure that you know exactly what you want to do before starting the array again. Do not take my thoughts below as a recommendation. Consider them and input from others and decide for yourself the next course of action. Recommendation #4. Update the BIOS. That would be an easy thing to fix and eliminate from being the problem. Every indication was that the 1TB drive is bad. But now you are saying that you have ANOTHER SATA drive failing to be recognized during your POST. Either you're unlucky, just happen to be stroring your hard diisks next to a large electromagnet, or there is another problem. Maybe that one port on the MB has a problem. I already suggested changing the cable. If you haven't, do that. Once you figure out what disks are good and what disks are bad, I think that what you will need to do is the following. Put your original disk back into slot 5, hit the "restore" button (this will lose all your parity data), and then start the array. Parity will be rebuilt using the 5 data disks. Once that is complete, redo the update process with a known good drive. IF YOU DO NOT HAVE ALL YOUR DATA DISKS INTACT, DO NOT DO THIS. Good luck -Brian
March 2, 200818 yr also, as far as the syslog is concerned. Where exactly am I typing that command line in. On my server it says Tower login, I then logged in with user name and password, and typed the command line as you mentioned, but the syslog doesn't appear in my flash drive. Am I doing this correctly? Sorry I've never used a Linux system before. Just to make it clearer where the spaces do and do not go, here's an 'emphasized' version of the command: cp /var/log/syslog /boot/syslog.txt After typing this, you should see syslog.txt in the root of the flash drive. From a Windows computer, you can browse to \\tower\flash and see it. Brian, I don't think he was saying that another SATA disk is missing, I took it that the same 1TB drive caused errors again. It looks to me to be incompatible with that port. I don't know exactly what motherboard he has, but there are reports of problems with ports 5 and 6 on a few motherboards with 6 SATA ports. A BIOS upgrade *might* fix this and the other errors, and I agree with Brian about the need to upgrade this one.
March 2, 200818 yr Author On my atual server I"m at the Twoer login: I then see the following: Tower login: (I type "Root") Password: (I type my password) [Disconnect bypassed -- root login allowed.] Linux 2.6.22.5. root@Tower:~# I then type cp /var/log/syslog /boot/syslog.txt and press enter it says root@Tower:~# When I browe to \\Tower\flash, there is a single folder "config" which contains "user" I don't see a syslog.txt file What am I doing wrong? Thanks again
March 2, 200818 yr You are probably doing fine... you probably have your windows machine set to hide system and hidden files. To see the file on the flash drive, you will need to set your windows file explorer to see system and hidden files. Open windows file explorer, under tools on its menu you will see options on how to view the data. On Vista it is "Tools" then "File Options" then click on the "View" tab, and lastly, click on the option to "Show Hidden and System files" and un-check the choice that hides file extensions for known file types. Odds are you will then be able to see the syslog.txt file. Joe L.
March 2, 200818 yr Ok, managed to catch my array as it was nearly done rebuilding my array. As you can see, my smallest drive, a 250 Gig is no longer spinning and has gone to sleep. That is because it is no longer being read to rebuild the new replacement disk2. Only 9.5 minutes to go from when I took this screen capture... Waiting a bit longer, finally, it is done. When I again press the "Refresh" button I now see this: Parity is now valid, all drives are on-line, and the new disk2 has 0 blocks read from it, and lots written to it. Disk3 is still sleeping since it has been over an hour since it was read from. Joe L.
March 3, 200818 yr Author Ok cool. Joe you were right, I needed to uncheck the "Hide protected operating system files" box. That ended up being the culprit. So attached is the syslog! Yay finally a step forward Brian, thanks again, I'll have to wait to flash the bios. All my USB drives are at work. I did replace the cable on "SATA 2" and it doesn't seem to make a difference. So right now I"m in the middle of a parity re-sync to get my system back to square one. Then my plan is to update the bios using Brian's method with the USB stick (I'm sure that will run smoothly, right ). Then I'll try one of the other SATA drives I have instead of the "buggy" 1TB drive and see if that upgrade works. Again I"ll have an extra 8 SATA ports from a PCI card in a few days which will make an upgrade (hard drive swap) unnecessary - but at this point I want to see this through so that I can have faith in my system's ability to recover from a dead drive. At least I'm working out the bugs when my data isn't truly at risk! Thanks again for all your help! Let me know what you think about the syslog! Ryan
March 3, 200818 yr Recommendation #1. TAKE YOUR TIME. Recommendation #2. If you ask for help here, make sure you post the diagnostic materials requested. Recommendation #3. Make sure that you know exactly what you want to do before starting the array again. Do not take my thoughts below as a recommendation. Consider them and input from others and decide for yourself the next course of action. <snip> Once you figure out what disks are good and what disks are bad, I think that what you will need to do is the following. Put your original disk back into slot 5, hit the "restore" button (this will lose all your parity data), and then start the array. Parity will be rebuilt using the 5 data disks. Once that is complete, redo the update process with a known good drive. IF YOU DO NOT HAVE ALL YOUR DATA DISKS INTACT, DO NOT DO THIS. Only use the "Restore" button if the drive you put back into slot 5 was the drive you took out to replace with the 1T drive AND it has not been in the array since you removed it. If it was there when you previously used "Restore" the odds are it is now empty and YOU DO NOT WANT TO REBUILD PARITY AS IT WILL LOSE ALL DATA THAT WAS ON DISK5. So, until we are sure you will not lose data, do not use the "Restore" button. It can ONLY be used to permanently REMOVE a drive and its data with the array in the state you have it currently. Let's take this one step at a time. Your last screen shot showed a smaller disk5 and the unRaid array complaining it is too small to be used as a replacement... First step is to put back in your original disk5 and get your array back to where it was before you started doing this upgrade of disks. If that drive is not available, un-assign disk 5 and bring the array back on-line. with it un-assigned, it should show up as "red" but you should still be able to get to its contents if your parity data is still valid. Get to this point and report back. If you can see the files on the un-assigned drive, copy them to one of your other drives to make sure you do not lose them. If you can do this, you no longer need to worry about re-building parity since you have a full set of data on your drives. Everything you have described points to a hardware issue. It could be the new 1T drive, or the disk controller, disk cable, or the power supply, or memory, or excessive heat (you do have a CPU heatsink... right?) I personally doubt if it is the bios, since it saw the drive when you first attempted the upgrade. (It might still benefit from an upgrade of the bios, but for now... let's not complicate matters) So... first step... get your data off of the old disk2, or read it from the unRaid array parity/data drives with disk2 un-assigned. Joe L.
March 3, 200818 yr ... If I reboot with the old 500gb drive in place, it boots up just fine. I've tried this with another SATA drive other than the "offending" 1TB drive and it too is not recongnized during the boot sequence - but it doesn't freeze. Unraid boots, but the disk is not recognized. Robj - This is what lead me to believe that there was another disk that was also not working. Your port 5/6 theory seems very likely here, as it appears that two different big drives are failing during POST on the same port. At first I thought it might be a BIOS issue with supporting the big disk, but he is running the exact same drive as parity so ruled that out. I didn't think about a different port being on a different controller. rpf717 - If the problem is with SATA Ports 5/6, and the BIOS update doesn't fix it, then you should be able to shufflle around the drives so that this new 1T is on one of the first four ports. Need to be careful, however, that whatever is put on the last 2 ports IS compatible, otherwise it could get clobbered. !!!UPDATE YOU BIOS FIRST!!! That may fix this whole mess. I would definitely run the smartctl on the drive if it works. It could be that being on the bad port and having unRAID writing to it could have caused sector remapping to occur that used up all your spare sectors. That would be bad! I have an idea how to fix it if that happened. Could you post your motherboard model?
March 3, 200818 yr Author Here is my equipment for the system, all of which was purchased new this month: MB: Asus P5B-VM DO Memory: 2 x 1GB Corsair XMS DDR 800 PSU: Corsair 550 watt model CPU: Celeron 440 2.0 ghz (with the standard heat sink installed) Definitely don't think heat is a problem, its in a thermaltake armor case and has plenty of cooling, my drives are typically 25-30 deg C The original 500gb drive is back in SATA 2 motherboard slot (drive 5 in the Unraid) - So its not in the SATA 5/6 slots - It is in the middle of a parity sync because when I put the drive back in the original configuration, the array would not start because it said the drive wasn't big enough (after taking the 1 TB drive back out). So I checked and all my data is there, I can read and write to the 500gb drive (Disk 5 in the array - SATA 2 on the motherboard).
Archived
This topic is now archived and is closed to further replies.