kimguan

Hopefully, this is in the correct forum; wasn't sure if I should have posted in General Support instead. My parity HDD is an Ironwolf 4TB; it survived 3 pre-clears with no errors. However, after the parity build, a notification popped up to say that the disk had 8 reallocated sectors. From reading other posts in the Forum, I understand that reallocated sectors should not be a problem unless the count begins increasing. However, that advice was always provided for data disks, not parity. The count on my parity drive has not changed since it was purchased and installed in January 2018. I suppose my question is: should I do anything about the parity disk with reallocated sectors, or is it all cool unless the count starts increasing (as with data disks)? I'm wondering if it would be "safer" to swap it out for use as a data disk instead? I'm about to add a second parity disk. Does this change anything with regard to any advice concerning reallocated sectors on parity disks? I'm wondering if it would be "better" if the HDD with the reallocated sectors were to be used for the second parity disk instead of the first? Sorry if this is pedantic on my part, but I tend to be OCD about these things... ?

This is interesting! This would imply that the parity algorithm is tied to the slot and not the number of disks. Slot 1 uses XOR and Slot 2 uses Reed-Solomon? Going off on a tangent now: out of curiosity, why was the decision made to limit unRAID TO 2 parity drives? Reed-Solomon should be able to support an arbitrary number. http://web.eecs.utk.edu/~plank/plank/papers/CS-96-332.pdf

Thanks that makes perfect sense. Thank you everyone for your help!

So the procedure would be: shut down server; physically rearrange disks; start server; stop array; go to Tools > New Config; "Preserve current assignments": none; select "I want to do this"; click "Apply"; Go to Main; Assign disks to slots; select "Parity is valid"; start array. Is that correct?

Ok - that makes it clearer. If the GUI does not automatically rearrange the order of the disks to reflect the physical location of the disks, can I do it manually without losing data or going through a parity rebuild? For example, if I physically swap the location of 2 data disks, can I stop the array, then unassign the 2 disks in the GUI and then reassign them to match their physical locations?

Can I clarify: If we start with this: Physical disks: (1) Data 2TB (2) Data 4TB (3) Parity GUI (matches physical arrangement): (1) Data 2TB (2) Data 4TB (3) Parity And I physically rearrange the physical disks thusly: (1) Parity (2) Data 4TB (3) Data 2TB Will the GUI show: Option 1 (matching original physical arrangement): (1) Data 2TB (2) Data 4TB (3) Parity Option 2 (matching new physical arrangement): (1) Parity (2) Data 4TB (3) Data 2TB Option 3: Some other arrangment If it is not Option 2, is there a way to change the GUI to match the new physical arrangement? Thanks in advance for your help!

Hi there - my setup has 7 x data, 1 x parity, 1 x cache disks. These disks are split between motherboard SATA and an LSI9201-8i card in no logical arrangement. 1. If I "tidied up" and physically rearranged the disks in the case, moving disks between the onboard SATA and the LSI card, would that cause any problems for the array? 2. Would the arrangement of disks in the web GUI update to reflect the new physical arrangement of disks in the case? Thanks!

Thanks trul! Looks like I'll have to reconsider my "solution" to this...

I am thinking of using 2 x 2TB HDDs and 1 x 120GB SSD in the default RAID1 configuration in a cache pool. The aim is three-fold: 1. to avoid Plex spinning up the array when not actively being used; 2. to provide a large download area for Transmission off the array; and 3. to provide some redundancy against failure of a disk. Does my proposed cache setup meet those goals? I guess my big assumption and hope is that disks in a Btrfs RAID1 cache pool are able to spin up and down independently, but this may not be possible because of striping? If my assumption is wrong, then would forgoing the SSD (which I have to buy) and just living with constantly spun-up 2 x 2TB HDDs (which I already have on hand) be the way to go?