bdistin

Following to see what you get for the JBOD. I have a bunch more 2.5 slots on my r730xd, but some exos 20tb drives could be pretty sweet someday.

Found a pretty cool docker app. Could be used to turn your HexOS server into a Steam Remote Play host. https://github.com/Steam-Headless/docker-steam-headless I would totally stream from my server to my SteamDeck handheld when the handheld GPU isn't cutting it.

I thought it was a very good deal for the refurbished unit. $399 before tax, came with 4 1.2tb sas drives/the cpus/ram, and it had lots of room to grow. All the rest of the budget was just adding more drives/drive sleds and the GPU. I knew I was migrating a decent sized plex library, wanted some ssd storage for apps, and a GPU that fit in the server case for transcoding (the 3050 is completely PCI bus powered, so no extra cables needed).

Background: I work as a System Administrator professionally managing Windows Servers, Virtualization, QNAP Nas's, and other network engineering. I had an old Windows Server I cobbled together at home years ago I was running plex, home assistant, and some other stuff; but it's been on its deathbed for a few years now. It was time for an upgrade, and LTT convinced me to give HexOS a try. The Build: Amazon Refurbished Dell R730XD w/ 128GB of ECC memory, dual Intel(R) Xeon(R) CPU E5-2690 v4 @ 2.60GHz (28-cores total), 6 - 1.2TB SAS HDD (going to add 2 more soon), 4 - 1TB SATA SSDs, OS on Mirrored 240GB SATA SSDs, NVIDIA 3050 for Hardware Transcoding. I have another 12 open drive slots to expand later if needed. Total cost of build: $1,272.82 The Configuration: I did a lot of stuff in unrecommended ways to see just how well HexOS/TrueNAS adapts, and went beyond a curated experience to get a good feel of the value HexOS adds to TrueNAS. This is also partially because I was replacing services I use with yet-to-be-curated apps. I have to say, HexOS handled what I threw at it very well. I found setting up a HomeAssistant VM in TrueNAS to be pretty straightforward, albeit more advanced than I would expect a family member to achieve, but not bad. I did install plex/immich via the curated way. Both required a touch of tweaks to get them to use hardware transcoding and a permission fix for images to get from uploads to the pictures folder, but they were technically usable in the curated state. Romm was another story though... I went through and made up the datasets w/ ACLs in the way plex/immich did, and doing that work manually is honestly an unintuitive pain. That's where my title comes from because I had begun to doubt how much HexOS was doing in the background to provide a polished experience. Getting Romm working abolished my budding doubts. Keep up the good work HexOS team!

I didn't mean that app deployment should move ahead of planned critical features and bug fixes. I was just suggesting that the current community could help with apps if the app-specific phases were reordered. Not that the average person would install "tons of apps" either. But some very good apps are being worked around to get going before the proper HexOS way or curated deployment is out. HomeAssistant in particular stands out to me, as I am guilty of getting a VM of it running through TrueNAS myself, but there are many docker containers/apps that would be appealing to other regular people such as mealie, gramps-web, romM, actual, etc.

Forgive me if this is too meta for this forum section. I had a few feature requests I wanted to write up. However, while doing my due diligence before posting, I found both were suggested but not yet supported in TrueNAS. I assume TrueNAS prioritizes more enterprise-impactful features because those are their paying clients. But now that there is a consumer revenue stream through HexOS, is there a path for consumer-friendly features to have a higher weight of priority? The requests I wanted to write up: WebUI File Manager https://ixsystems.atlassian.net/browse/NAS-109580 GPU info Widget https://github.com/truenas/webui/discussions/9865

This may not be the right place to make this argument, and I fully understand your plans may be firm and I am wasting my time by saying this. However, I would argue you almost certainly have a captive base of enthusiasts beta-testing your product rather than a high percentage of normies. (I apologize for that term; I can't think of a better word and don't mean it as an insult to anyone.) Given that assumption, it would seem that the phases are almost backward in an ideal world. Enthusiasts are likely going to go the extra distance to get what they want running on their NAS whether they have a HexOS way of doing it, or if they have to do it from TrueNAS. It would make more sense to have the custom docker/deployment script loader available first, with a suitable warning of dragons and dangers ahead, and provide an open-source example of a curated script so the community can start making and sharing custom app deployment scripts the HexOS way. Maybe have a forum section for sharing and collaborating/improving the scripts. Then you would already have a portfolio of proven community developers once you are ready to recruit, and (if you so choose) you can give out a "HexOS stamp of approval" to well-designed community app deployments. Finally, for the popular yet difficult apps that require a demanding configuration/deployment, work on curating them yourself later. I understand that you might need to make a few curated apps to define the "HexOS" way of app deployment and to develop a best practices guide for the community before allowing custom deployments, but by just opening the door, I suspect you will have a very nice catalog before you know it. Furthermore, I suspect the community will naturally develop deployments for popular apps you may already be internally planning on curating, therefore, saving time and money from having to develop them yourself. This would ultimately allow more development budget for critical features/integrations. To be clear, if HexOS were in wide release, I would completely agree with your planned phases. However, because this is still a beta, I think it would be advantageous for HexOS to foster community development sooner rather than later.