December 7, 201312 yr So what is the process to replace a smaller disk with a larger one? I want to remove a 250 gig drive and replace it with a 3 TB.
December 7, 201312 yr This should work http://lime-technology.com/wiki/index.php/Troubleshooting#Replace_the_drive is a bit old but i think should work haven't done it in a LONG while..
December 7, 201312 yr shut down array shut down system take out 250gig put in 3tb start system start array system will tell you a disk is missing, you can then assign the new disk. parity will then reconstruct the drive.
December 7, 201312 yr It seems on V5 with replacement disks that once you start the array you have to press a button to start the build or rebuild otherwise the array will start and just stay in a degraded state.
December 16, 201312 yr Author shut down array shut down system take out 250gig put in 3tb start system start array system will tell you a disk is missing, you can then assign the new disk. parity will then reconstruct the drive. thanks, seems to be working. rebuilding it now
December 16, 201312 yr Ok,there's something i'm not getting, as it almost looks like magic.Its always recommended to have a parity disk as big or bigger than the largest drive in the array. SO, I might have 20TB array which comprises of 5 4TB drives, and then have a 4TB parity drive. I might then remove disk A, and then the parity drive will rebuild disk A completely? but I might wanna replace disk B, and not disk A, and somehow the parity disk will rebuild disk B?? I mean, how is this possible without us telling unRAID first before removing the drive what we are about to do, so than maybe the parity drive will make sure all the info of the about to be removed drive is in it, and only than allow us to remove the drive? sorry, but it really seems like magic now, so hopefully someone would be able to explain a little. Thank you in advance.
December 16, 201312 yr Ok,there's something i'm not getting, as it almost looks like magic.Its always recommended to have a parity disk as big or bigger than the largest drive in the array. SO, I might have 20TB array which comprises of 5 4TB drives, and then have a 4TB parity drive. I might then remove disk A, and then the parity drive will rebuild disk A completely? but I might wanna replace disk B, and not disk A, and somehow the parity disk will rebuild disk B?? I mean, how is this possible without us telling unRAID first before removing the drive what we are about to do, so than maybe the parity drive will make sure all the info of the about to be removed drive is in it, and only than allow us to remove the drive? sorry, but it really seems like magic now, so hopefully someone would be able to explain a little. Thank you in advance. How does parity work?
December 16, 201312 yr Its always recommended to have a parity disk as big or bigger than the largest drive in the array. It's not recommended, it's MANDATORY.
December 17, 201312 yr Parity: Parity is like a 'checksum', its an error correction mechanism...you see checksums every day. There's one built into the number on the credit card you used for lunch today. Here's a simplified concept: Assume 3 data drives + Parity. Drive 1's data = 1,2,3,4 Drive 2's data = 4,4,4,4 Drive 3's data = 9,9,8,8 Assume the parity algorithm is as follows: Add up all the data, 'by column' and write it to Parity So, column 1 has a 1, a 4 and a 9 = 14 column 2 is 2,4,9=15 column 3 is 3,4,8=15 column 4 is 4,4,8=16 Apply the algorithm to get the parity drive: So the parity drive's columns of data = 14,15,15,16 IMAGINE now that Drive 1 is crashed/borked/removed...you still have Drives, 2,3, and parity. Install a new Drive 1. Run your trusty unRAID server and it will recreate Drive 1 by taking the Parity Drive, and subtracting the values from the other Drives (Drives 2 and 3) So the new Drive 1 Column 1 Gets: 14 (from Parity) - 4 (from Drive 2) - 9 (from Drive 3) = 1 UNRAID Writes '1' in 'column 1' on Drive 1. Column 2 Gets: 15-4-9=2 UNRAID Writes '2' on 'column 2' on Drive 1. Column 3 Gets: 15-4-8=3 , etc. Column 4 Gets: 16-4-8=4 The rebuilt Drive 1 now has the data "1,2,3,4". This same algorithm works for any data drive, not just 'Drive 1', as long as you are replacing or 'error correcting' only one drive at a time. You can recreate any ONE single drive failure or replacement, but it requires the other drives work and be errorfree. That's why its important to follow the instructions for replacing/repairing drives, etc. You want to avoid situations where you have multiple broken drives...the algorithm can only recover one drive at a time. http://en.wikipedia.org/wiki/Parity_bit
December 17, 201312 yr Parity: Parity is like a 'checksum', its an error correction mechanism...you see checksums every day. There's one built into the number on the credit card you used for lunch today. Here's a simplified concept: Assume 3 data drives + Parity. Drive 1's data = 1,2,3,4 Drive 2's data = 4,4,4,4 Drive 3's data = 9,9,8,8 Assume the parity algorithm is as follows: Add up all the data, 'by column' and write it to Parity So, column 1 has a 1, a 4 and a 9 = 14 column 2 is 2,4,9=15 column 3 is 3,4,8=15 column 4 is 4,4,8=16 Apply the algorithm to get the parity drive: So the parity drive's columns of data = 14,15,15,16 IMAGINE now that Drive 1 is crashed/borked/removed...you still have Drives, 2,3, and parity. Install a new Drive 1. Run your trusty unRAID server and it will recreate Drive 1 by taking the Parity Drive, and subtracting the values from the other Drives (Drives 2 and 3) So the new Drive 1 Column 1 Gets: 14 (from Parity) - 4 (from Drive 2) - 9 (from Drive 3) = 1 UNRAID Writes '1' in 'column 1' on Drive 1. Column 2 Gets: 15-4-9=2 UNRAID Writes '2' on 'column 2' on Drive 1. Column 3 Gets: 15-4-8=3 , etc. Column 4 Gets: 16-4-8=4 The rebuilt Drive 1 now has the data "1,2,3,4". This same algorithm works for any data drive, not just 'Drive 1', as long as you are replacing or 'error correcting' only one drive at a time. You can recreate any ONE single drive failure or replacement, but it requires the other drives work and be errorfree. That's why its important to follow the instructions for replacing/repairing drives, etc. You want to avoid situations where you have multiple broken drives...the algorithm can only recover one drive at a time. http://en.wikipedia.org/wiki/Parity_bit So that's how it works! suddenly it all makes sense now in terms of how a parity drive is able to rebuild a drive in the event of a single disk failure. Thanks Dale for the clear and concise explanation.
Archived
This topic is now archived and is closed to further replies.