September 21, 20223 yr I have a unraid box with 6.9 running with 2 6tb parity drives and 9 6 or 4 tb array drives. One of the 4 tb drives failed and is disabled with many read failures. I bought 10 TB drive to replace one of the parity drives then move it to the dead array drive. What is the safest way of doing this ? Can they be done at the same and is that safer than : Moving the parity drive to replace the array drive and rebuild Then install new party drive . John
September 21, 20223 yr Community Expert Solution You can use the Parity Swap procedure that is designed for exactly this Use Case.
September 21, 20223 yr Community Expert Might be a good idea to post diagnostics so we can see if there is any indication why the disk became disabled. Connection problems are much more common than bad disks.
September 21, 20223 yr Author Thank you for your help That was the link i was looking for but could not find. I looked at the smart diag on the drives page and it has many sectors that are pending and unable to then the disk was disable on a unable to read error.
September 21, 20223 yr Author This procedure looks safe as parity is maintained the whole time (the reason for the copy). just time consuming. The array will be offline for the copy process. So a 10 tb replaced 6 tb party then then 6tb party replaces dead drive on second step on basic rebuilt drive operation.
September 24, 20223 yr Author One part of the instructions is missing or i missed it. Do not shutdown after the copy phase and before the rebuild disk part I did that and i forgot or did not remember the copy phase. So i have to do the copy phase again ( 14 hours) on a 10 TB disk. I moved the drives physically in the system after the copy part.
September 25, 20223 yr Author New problem with the replacement of drives. The copy was successful but drive 2 which was not involved in the procedure shows wrong or no file system. The system is rebuilding but 500k bytes a second. What do i do next ? tower-diagnostics-20220925-0747.zip
September 26, 20223 yr Author New problem when doing a xfsrepair the log get full and it stops Metadata corruption detected at 0x44e910, xfs_bmbt block 0x28d2ea458/0x1000 libxfs_bwrite: write verifier failed on xfs_bmbt bno 0x28d2ea458/0x8 Metadata corruption detected at 0x44e910, xfs_bmbt block 0x28d079b50/0x1000 libxfs_bwrite: write verifier failed on xfs_bmbt bno 0x28d079b50/0x8 Maximum metadata LSN (1095979799:38281785) is ahead of log (1:2). Format log to cycle 1095979802 How do i go forward from here?
September 26, 20223 yr Community Expert There are issues with multiple disks, could be a power/connection problem or just the controller.
September 26, 20223 yr Author the log was full. It finished but needed to be run a second time to complete. The disk is still showing disabled tower-diagnostics-20220926-0652.zip
September 27, 20223 yr Author There were loose cables for power to the top drive cage. I corrected that and have started a rebuild. Power issues cause strange problems with file system errors that i had not seen before. I did not even touch these recently.
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.