I've been reading everything I can about ZFS and watching the videos from spaceinvader and others, but Im a little confused about things and what to do!

Im building a new server to replace an old one I cant expand any more.  So I was going to take advantage of ZFS.

I will be using 8x 4TB  drives, ie they are the same capacity.  In doing this, am I better off doing RAIDZ2 or XFS with 2 parity?

 

ZFS gives me compression which gives more capacity; I should also get higher read/write speed (I dont have CPU issues as it is a new server dedicated to running Apps in dockers); I can do snapshots to a backup machine.  

Im assuming that using RaidZ1 means the parity is spread across the drives so I dont have the performance dependant on the single parity used by Unraid, this will also mean less wear on a single drive used for parity due to less read/writes on the single drive.

 

XFS gives me easy capacity increase (but Im at max capacity in an N2 case with 8 drives) and less power as I only spin up the drive I am using.; less heat as Im in a small case with limited cooling;

 

What I am after is the highest resilience and good performance; I dont care about power consumption or processing impact. 

 

I naturally think I want to use RaidZ2.  But is this correct? Should I build the new box as a pure ZFS with RaidZ2, or as XFS with 2 parity, or some hybrid like ZFS striping with 2 parity, or RaidZ1 and 1 parity?

Another option would be 2 pools of 4 drive RaidZ1, so only half the drives spin up at any time when using them.  This then gives both speed benefits and also power and heat benefits.

 

I should also add I will have a 1TB M.2 for the cache.

 

what is the reason not to use ZFS? and what config do people suggest and why?  (the why is important!)

 

Its all new on UNRAID, and parity vs zfs makes it a little complex in the choices, and I can see loads of opportunities to get it right or make a pigs ear of the initial config!

 

Thoughts please.

Edited January 15, 20242 yr by tigga69

If you do an all ZFS pool, you would need to add 1 drive to the array. Could be a USB drive.

One other option is a 2 parity but all ZFS array(all individual disks). That would give you the compression, not have to spin up all the disks, but you would lose the self-repair option of a ZFS Pool.

HI Foo_fighter...

not sure I understand your response.  

On 1/16/2024 at 6:52 AM, foo_fighter said:

you would need to add 1 drive to the array

 

Why????  What size??? How should this be configured? Is this for Unraid as parity, or ZFS pool? 

If I use 6+2 for ZFS, then why would I need an extra drive.   I have an M.2 for the cache.  Also I still have 2 spare SADA ports, so I can add more drives, but why would I?

 

As I said Im after the highest resiliance, so no I dont want to loose the self-repair.  The whole point is maximum resiliance which is erring to ZFS as RaidZ2.  ie 6+2 in the pool.  

 

Option1: ZFS with RaidZ2 - 6+2 drives:  Pros:self repair, performance  Cons: high power usage, high heat from drives, high wear as all drives spin

Option2: ZFS like normal unraid - 6ZFS single drive pools + 2 parity:  Pros: low power, heat, and wear.  Cons: no self repair, slow performance.  

Option3: ZFS with Raid0 - 6 drives striped ZFS pool without fault tollerance but with 2x Unraid paritydrives:  I suggested this but it wouldnt work as the ZFS pool is larger than the parity drives. can only be done as option 2 unless large parity drives purchased.  

Option4: 2x RaidZ1 pools - 4 4x drive pools, joined in Unraid without parity ie 3+1 ZFS pool & 3+1 ZFS pool, Pros, stripping across 4 drives improved performance over option 2, maintains self repair, lower drive wear, lower heat, lower power (less drives spining). Cons: lower performance, lower resilience (1 drive failure).

 

As I write it this way, I am still convinced it is either option 1 or option 4.  Option 2 doesnt makes sense for my use case. (and option 3 doesnt work)

 

You appear to be suggesting either I configuration I dont understand, or option2 which doesnt meet my needs.  This is just a variant on today's setup with XFS but using ZFS with minimal benefit.

 

 

 

26 minutes ago, tigga69 said:

HI Foo_fighter...

not sure I understand your response.  

 

Why????  What size??? How should this be configured? Is this for Unraid as parity, or ZFS pool? 

If I use 6+2 for ZFS, then why would I need an extra drive.   I have an M.2 for the cache.  Also I still have 2 spare SADA ports, so I can add more drives, but why would I?

Currently unRAID requires there to be at least one drive in the "Array" (meaning the disk pool that uses the unRAID driver), regardless of what other pools are specified. For people who want to use only ZFS pools and no unRAID "Array" the workaround is to assign a single small USB flash drive (other than the boot drive) to the array to satisfy the requirement. Unfortunately the drive does still count against drive limit for the particular unRAID license level. 

It is expected that this requirement will be going away at some point as Limetech makes further improvements for supporting drive pools. Unfortunately, as is the case with most unRAID development, there is no ETA for that feature.

Edited January 21, 20242 yr by primeval_god

ok thanks thats clearer

 

So I build a single drive array, a cache, and a zfs pool (which has most of my storeage capacity).

I've the Plus licenses so Im ok with the drive numbers and oddly I have a spare small SSD lying around, so that's ok too.

 

I'll be building the system this weekend (I think) so I can see what it all looks like.

 

Again thanks for the clarity.

 

Edited January 24, 20242 yr by tigga69

Digging up this post to ask about my options for possibly converting to ZFS, especially with the v7 ZFS additions.

I currently have an 88Tb array, with 2x12Tb parity. My array is a combination of drive sizes - 7x4Tb, 1x8Tb, 4x10Tb and 1x12Tb. My current usage is 73.5Tb out of 88Tb total. My array is almost exclusively for media files, so I don't know if compression would benefit me with the majority of files converted to h265. I'm planning on ordering another 12Tb drive this weekend, as well.

So, with my various drive sizes, what are my options other than ordering enough 12Tb drives to hold everything? I do want to use a RaidZ, either level 1 or 2, and would like for the drives to be able to spin down when not in use. I'm completely new with ZFS, with just created 2 work pools as basic Raid-0.

8 hours ago, malaki86 said:

So, with my various drive sizes, what are my options other than ordering enough 12Tb drives to hold everything?

The only other option is to use the current drives with single or dual vdevs, but that would waste a lot if space, especially the single vdev option. 

 

8 hours ago, malaki86 said:

and would like for the drives to be able to spin down when not in use.

Zfs pools can still spin down, but it's all or none, unlike the array, i.e., reading a single file will spin up all the disks.

Is the current state of ZFS at a stable and future-proof point where I can create a RAIDZ2 vdev (alone in a pool) out of 6 (same capacity) drives - and then when 7 (and later) are released, be able to take advantage of the hybrid solid-state cache feature(s) without having to destroy or remake the vdev and pool?

 

How about later when OpenZFS 2.3 is integrated with the ability to expand vdevs? I know current vdevs on TrueNAS for example are OK, I just want to make sure Unraid's will also be good to go.

Edited June 29, 20242 yr by Espressomatic

14 hours ago, Espressomatic said:

Is the current state of ZFS at a stable and future-proof point where I can create a RAIDZ2 vdev (alone in a pool) out of 6 (same capacity) drives - and then when 7 (and later) are released, be able to take advantage of the hybrid solid-state cache feature(s) without having to destroy or remake the vdev and pool?

Yes.

 

14 hours ago, Espressomatic said:

How about later when OpenZFS 2.3 is integrated with the ability to expand vdevs?

Should also be fine. 

8 hours ago, JorgeB said:

Should also be fine. 

 

OK, that's good to know, as I don't want to have to do a major data shuffle again later in the year. - EDIT: I decided to jump right onto v7 beta1 to set this up.

 

This time around I used a great unix/linux compression utility (rm) that can take any amount of data and squeeze it down to 0 bytes. 

Edited June 30, 20242 yr by Espressomatic

Regarding Unraid 7.0 and ZFS

 

During my test setup usage I realised I'm not happy with regular ZFS array (not ZFS pool) read/write speeds (4 x 6 TB WD Red Plus). Unraid 7.0 brings L2ARC and SLOG subpools, but will they be useable for ZFS arrays (not pools)? In hybrid ZFS array you can have pools and datasets, but is there any info that L2ARC and SLOG will work too?

2 hours ago, YujiTFD said:

Unraid 7.0 brings L2ARC and SLOG subpools, but will they be useable for ZFS arrays (not pools)?

Nope, array zfs devices can only be a simple single device pool.

2 hours ago, JorgeB said:

array zfs devices can only be a simple single device pool

Does this mean if I choose a ZFS pool over a ZFS array, I lose all the array benefits? For example, won't I be able to add more HDDs into the pool as easily as into the array?

After I set up 5 HDDs as RAID-Z1, can I add more drives later, and can these drives be of different sizes?

 

I had a full-blown TrueNAS SCALE setup (3 x mirrored vdevs). While I/O and transfer speeds were awesome, I had no use for such a system, and about 85% of the time it was idling, with occasional movie access and very modest torrenting. It felt like a waste of hardware and electricity bills, so I went for Unraid for its perks, finding it quite slow in comparison. I can hardly saturate a 1G network, let alone the 10G network I had installed, I'm looking for improvement without sacrificing too much >_<

29 minutes ago, YujiTFD said:

Does this mean if I choose a ZFS pool over a ZFS array, I lose all the array benefits?

See if this answers your question:

https://forums.unraid.net/topic/131857-soon™️-612-series/?do=findComment&comment=1243850

 

So basicly we can have either one or another, something hybrid with best of both worlds are not possible?

We can somewhat mitigate writing speed with SSD cache, but nothing for reading? Shame.

You can't have everything.