JorgeB Posted November 20, 2019 Share Posted November 20, 2019 Disk2 will remain as it was after a new config, i.e., with the data it was after you ran xfs_repair. 1 Quote Link to comment
Bedey Posted November 20, 2019 Author Share Posted November 20, 2019 Ok no problem, not a big loss, by the way last time I tried to put a new hard drive I was following the wiki and I didn't do this step 9.Put a check in the Yes, I'm sure checkbox (next to the information indicating the drive will be rebuilt), and click the Start button I didn't see any chechbox with this option, only one saying to format the drive(I didn't clicked it), I missed something? Next week I'll try to add again my bigger drive and I want to be sure I do everything correctly Quote Link to comment
JorgeB Posted November 20, 2019 Share Posted November 20, 2019 13 minutes ago, Bedey said: I didn't see any chechbox with this option There can be little differences from version to version, wiki isn't always up to date, mount problems are unrelated to that. Quote Link to comment
Bedey Posted November 22, 2019 Author Share Posted November 22, 2019 Ok, today I tried to put back the new 3tb drive, everything went well until the end of the rebuild, the new drive give me the same error (unmountable :no file system) so I stopped the array to try to fix the file system, but for no reason my drive have a different name than before, how I should procede? Quote Link to comment
JorgeB Posted November 22, 2019 Share Posted November 22, 2019 Device can't be on a controller that changes identification, Unraid requires disks identification to remain the same, or it will consider it a different disk. 1 Quote Link to comment
Bedey Posted November 22, 2019 Author Share Posted November 22, 2019 The new drive is connected to the motherboard, but it changed the name after I stopped the array, i have to create a new configuration? Or I have to start the array in maintenance mode and fix it's file system Quote Link to comment
JorgeB Posted November 22, 2019 Share Posted November 22, 2019 21 minutes ago, Bedey said: The new drive is connected to the motherboard, but it changed the name after I stopped the array That shouldn't happen with the onboard ports, please post current diags. Quote Link to comment
Bedey Posted November 22, 2019 Author Share Posted November 22, 2019 I have two drives with the same ID, in fact the first time I plugged the new drive he got that weird name, but is not supposed to change after I stopped the array,right? petere-diagnostics-20191122-1011.zip Quote Link to comment
JorgeB Posted November 22, 2019 Share Posted November 22, 2019 I don't know what kind of device that is since there's not even a SMART report, but you don't want to be using that with Unraid, if it changes ID without warning it won't work. 1 Quote Link to comment
Bedey Posted November 22, 2019 Author Share Posted November 22, 2019 Is a 3tb drive like the one used on disk 2,so I need a new hard drives or what? Quote Link to comment
JorgeB Posted November 22, 2019 Share Posted November 22, 2019 Use standard hard drives, from WD (HGST), Toshiba or Seagate. Quote Link to comment
Bedey Posted November 22, 2019 Author Share Posted November 22, 2019 But is a 3tb Seagate hard drive, I can't understand why unraid is incompatible with a common hard drive Quote Link to comment
JorgeB Posted November 22, 2019 Share Posted November 22, 2019 It's not being detected as a standard hard drive, try to get and post a SMART report. Quote Link to comment
JorgeB Posted November 22, 2019 Share Posted November 22, 2019 (edited) OK, I see the problem now, disk2 has the same ID as disk3: Nov 21 18:06:38 Petere kernel: ata10.00: ATA-10: OOS3000G, 00000000, 0001OOS1, max UDMA/133 Nov 21 18:06:38 Petere kernel: ata9.00: ATA-10: OOS3000G, 00000000, 0001OOS1, max UDMA/133 You can only use one of these with Unraid, and these are no standard Seagate drives, maybe white label? Edited November 22, 2019 by johnnie.black Quote Link to comment
Bedey Posted November 22, 2019 Author Share Posted November 22, 2019 Yes, white label, is that the problem? Quote Link to comment
JorgeB Posted November 22, 2019 Share Posted November 22, 2019 In this case yes, since they have the exact same ID, they don't even have a valid serial number, Unraid requires that each device has a different ID, since that is used to track them. Device Model: OOS3000G Serial Number: 00000000 1 Quote Link to comment
Bedey Posted November 22, 2019 Author Share Posted November 22, 2019 Ok, thanks for your help Johnnie, I'll buy another HD and try again Quote Link to comment
Bedey Posted November 27, 2019 Author Share Posted November 27, 2019 (edited) Ok it's me, again.. I replaced the drive and started the rebuild, but it give again the same error "Unmountable :No file system" so I just run a check with -n and it said that the first superblock was missing, so I run again the repair tool with the -v, but apparently it just recovered some files under the "lost-found" folder, I did something wrong? Edited November 27, 2019 by Bedey Grammar Quote Link to comment
JorgeB Posted November 27, 2019 Share Posted November 27, 2019 Like mentioned above it's not normal to get so many unmountable disks without an apparent reason, you likely have some hardware problem, like bad RAM, board, controller, etc., start by running memtest. Quote Link to comment
Bedey Posted November 27, 2019 Author Share Posted November 27, 2019 So after replacing a disk it shouldn't give me any error if everything is good? I can't understand this behavior, the nas was stable even after two months of consecutive running, I'll try with a memtest, unfortunately right now I don't have any spare ram or hardware to test or replace Quote Link to comment
JorgeB Posted November 27, 2019 Share Posted November 27, 2019 5 minutes ago, Bedey said: So after replacing a disk it shouldn't give me any error if everything is good? If parity is correctly in sync, even before replacing the disk while it's disabled or even unassigned, the disk is emulated and all data there should be available, the same data that shows on the emulated disk will be on the rebuilt disk. Sometimes filesystem gets corrupted when the disk is disabled, but unusual to corrupt the super block, and IIRC it's the third time it happens for you in a short space of time, so not normal at all, and filesystem corruption should never happen when upgrading disks, as long as parity is valid and in sync. 1 Quote Link to comment
Bedey Posted November 27, 2019 Author Share Posted November 27, 2019 Now it's all clear, however, disk 3 in now full of good files but with total different names(numeric sequence of numbers), can i still fix this? i think i'll order a new pc and move everything on the new one Quote Link to comment
JorgeB Posted November 27, 2019 Share Posted November 27, 2019 8 minutes ago, Bedey said: can i still fix this? Possibly, but you'd need to go through each file manually and rename. 1 Quote Link to comment
Bedey Posted November 27, 2019 Author Share Posted November 27, 2019 1 minute ago, johnnie.black said: Possibly, but you'd need to go through each file manually and rename. as i expected, well no problem, last question, moving all my drives in a new totally different system can cause any problem? like a ryzen computer Quote Link to comment
JorgeB Posted November 27, 2019 Share Posted November 27, 2019 Should be plug'n play for the data disks, only if RAID controllers are used there could be issues if they don't use the standard drive identifier string. Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.