May 26, 20215 yr I'm curious to know how if anyone out here uses Unraid for Davinci resolve database / collaboration purposes. I've managed to get postgres working with openVPN but not with wireguard for over the internet collaboration. I'm curious what other peoples setups are. There is someone working on a cool app to do this exact thing but inside windows / mac os https://github.com/jonnyhyman/ResolveCollaboration I attempted( By attempt I mean i don't know what I'm doing) to create a docker image from his project and oh boy is the boy is the learning curve steep when you don't know anything XD
January 21, 20224 yr @Wannabelinuxuser I've search and came cross your post, just generally on Davinci resolve and Unraid in general. I'd love to expand on this conversation, but haven't even started to explore. I just use shares for editing in DR in general but this is next level. I'm a lone editor so I dont need to do what you describe but it be a fun top to explore.
February 4, 20251 yr I'm a professional colorist and editor in the film industry. I work primarily on mac, and I'm trying to learn how to implement postgres sql databases on a linux workstation and my main mac workstation. I have an old PC I'm looking to convert into a server. I've been eyeing Unraid for this and I'd be happy to share my experiences once I actually build the server and test out Resolve across the systems. I do have one question regarding Unraid. It has a little set up: This is coming from my own lack of knowledge of servers and Unraid. I was thinking of running ZFS 4x4TB nvme's off one set of PCIE slots for VM distro testing, as well as holding my postgres sql databases for Da Vinci Resolve. In addition I was thinking of moving some movies and other media to a btrfs pool of several 8tb retired drives I have for a home plex server. Additionally a final pool of maybe zfs as an extra back up for all my work from my main work RAID. To summarize. 3 pools: 1. nvme pool using zfs for postgres sql databases for Da Vinci and running VMs for distro testing. Probably for caching. 2. HDD pool using ZFS as a back up storage for my main work RAID. 3. HDD pool using BTRFS for Plex. A point regarding RAM: I am using about 16GB of DDR3 ram that is non ECC (The motherboard does not support ECC) The question: can Unraid handle this sort of set up? Are there any specifics I can keep in mind? Am I missing anything here in terms of certain logic etc? Looking forward to hearing back. This is my first server, and I'm using it as a way to tinker, and learn and hopefully set up another server that's production ready down the line.
February 4, 20251 yr @Pabloicecreambar Honestly, it seems like you already have a solid understanding of what you need. For older hardware, Unraid is a perfect fit for your setup. By caching your DR database and VM on a designated drive, then backing them up daily to your RAID array, you’ll have two reliable copies. Unraid’s Unassigned Devices feature also makes it easy to maintain extra backups whenever you want. Another huge benefit is that only your parity drive needs to be the largest in the array, so you can mix different drive sizes to your heart’s content. It really does feel like Unraid was designed for exactly this purpose—especially if, like many editors, you’ve collected a variety of hard drives over the years.
February 4, 20251 yr Thank you for the response. Some potential issues I'm seeing now that I have some more information. 1. I think the potential bottleneck I'm hitting here with the unraid server I want to build is that the motherboard only supports DDR3 non ECC memory, and caps out at 32GB of RAM. Will this have issues with the ZFS volumes? 2. Regarding the NVME pool, one stick will have to be dedicated to being the cache for the ZFS pools. Is it possible to make that one NVME be the cache for both ZFS pools or is that not recommended?
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.