May 28May 28 This is my first time, so please be gentle.I'm setting up a homelab server, I have 3x18Tb HDD, 1x1Tb NVMe, 1x500Gb SATA SSDMy plan is to have 2xHDD for photos docs and sensitive data in an array (1 parity, 1 data)Then, 1xHDD for stored media (movies, music, TV shows... *arr + jellyfin)I asked advice for the LLM friends but they gave me different approaches, one told me to put downloaded cache and apps data (dockers, and 1VM for HomeAssistant, and jellyfin data) on the NVMe and use the SSD for caching backups that mover should later put into array...The other one told me to use NVMe for cache and SSD for apps, and then I question it about it and it inverted them, NVMe for apps and SSD for cache...And eventually they hallucinated, I believe, at least partially, due to the change on the mover options settings change from version >6.12? ... here is a scheme:here is a scheme:PARITY └── 18 TB HDD └── parity protection ARRAY (XFS) └── 18 TB HDD [protected sensitive data] ├── photos │ ├── Primary Storage: array │ ├── Secondary Storage: backuppool │ └── Mover Behaviour: Array → Cache │ ├── documents │ ├── Primary Storage: array │ ├── Secondary Storage: backuppool │ └── Mover Behaviour: Array → Cache │ ├── personal_backups │ ├── Primary Storage: array │ ├── Secondary Storage: backuppool │ └── Mover Behaviour: Array → Cache │ └── phone_uploads ├── Primary Storage: array ├── Secondary Storage: backuppool └── Mover Behaviour: Array → Cache MEDIA POOL (XFS) └── 18 TB HDD [unprotected media] ├── movies │ ├── Primary Storage: media │ ├── Secondary Storage: none │ └── Mover Behaviour: no mover │ ├── series │ ├── Primary Storage: media │ ├── Secondary Storage: none │ └── Mover Behaviour: no mover │ ├── music │ ├── Primary Storage: media │ ├── Secondary Storage: none │ └── Mover Behaviour: no mover │ └── downloads_archive ├── Primary Storage: media ├── Secondary Storage: none └── Mover Behaviour: no mover FASTPOOL (ZFS on NVMe 1 TB) └── Samsung 950 EVO NVMe ├── appdata │ ├── Primary Storage: fastpool │ ├── Secondary Storage: backuppool │ └── Mover Behaviour: Cache → Array DISABLED │ ├── system │ ├── Primary Storage: fastpool │ ├── Secondary Storage: none │ └── Mover Behaviour: no mover │ ├── domains │ ├── Primary Storage: fastpool │ ├── Secondary Storage: backuppool │ └── Mover Behaviour: Cache → Array DISABLED │ ├── downloads │ ├── Primary Storage: fastpool │ ├── Secondary Storage: media │ └── Mover Behaviour: Cache → Array │ ├── jellyfin_metadata │ ├── Primary Storage: fastpool │ ├── Secondary Storage: none │ └── Mover Behaviour: no mover │ ├── jellyfin_cache │ ├── Primary Storage: fastpool │ ├── Secondary Storage: none │ └── Mover Behaviour: no mover │ ├── transcodes │ ├── Primary Storage: fastpool │ ├── Secondary Storage: none │ └── Mover Behaviour: cache only │ └── unpacking_temp ├── Primary Storage: fastpool ├── Secondary Storage: none └── Mover Behaviour: cache only BACKUPPOOL (BTRFS on SATA SSD 500 GB) └── Samsung 850 EVO SATA SSD ├── appdata_backup │ ├── Primary Storage: backuppool │ ├── Secondary Storage: array │ └── Mover Behaviour: Cache → Array │ ├── HA_backup │ ├── Primary Storage: backuppool │ ├── Secondary Storage: array │ └── Mover Behaviour: Cache → Array │ ├── VM_backup │ ├── Primary Storage: backuppool │ ├── Secondary Storage: array │ └── Mover Behaviour: Cache → Array │ ├── snapshots │ ├── Primary Storage: backuppool │ ├── Secondary Storage: none │ └── Mover Behaviour: no mover │ └── emergency_restore ├── Primary Storage: backuppool ├── Secondary Storage: none └── Mover Behaviour: no movercould you please advise on the best way to set things up, poolwise, sharewise, filesystemwise.Thank you Edited May 31May 31 by jorgemarmo
May 28May 28 Community Expert There's no point in having a separate pool of a single HDD for "media" use.Either:Add all three HDDs to the unraid array 1 parity + 2 Data disksorCreate a single RaidZ1 zpool with all three disks.The caveat with option #2 is you would need to acquire another 18TB disk if you wish to expand the zpool however this can be done 1 disk at a time now for a single vdev. If you wanted to expand adding a 2nd vdev would require a same number of disks as the first vdev.
May 28May 28 Author I thought that leaving "media" out of the parity will :1- preserve (mechanically) the other two disks2- reduce CPU usage of calculating parity for all "media" dataStill, I don't know who to use as cache and how...
June 1Jun 1 Community Expert On 5/28/2026 at 10:06 AM, jorgemarmo said:I thought that leaving "media" out of the parity will :1- preserve (mechanically) the other two disks2- reduce CPU usage of calculating parity for all "media" dataStill, I don't know who to use as cache and how...You wont have enough disks in the array to assign parity. The pools have self-contained parity separate from the unraid array.Modern CPUs have no issue calculating parity, the limiting factor is raw disk read speed.I gave you the most optimal setup. Assign the SSDs to a pool, set it up as RAIDZ1(zfs) and then assign the single HDD to the Disk1 slot in the array.You do not have enough disks to really do anything else with.The reason you should not assign ssds to the array are due to it breaks trim and your ssds will degrade much faster without garbage collection.
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.