All Activity

Just sharing some good news — my update went smoothly, and all apps, VMs, and backups are running normally under Fangtooth.

Here's another point of confusion: Immich thinks the server version is out of date because it's comparing the TrueNAS-reported version rather than the release version, but this may be more of a bug for the TrueNAS team.

Thank you for the clarification! I think HexOS should be using the application version (2.2.3) in the UI, not the version (1.10.12). That way, if beginners need to Google problems, they can reference the right version. I've been using TrueNAS SCALE for about 5 years now, so I'd say I'm intermediate, but this one still tripped me up. 😅 I misspoke here and meant to say that TrueNAS tightly controls versions and community-maintained applications, not HexOS and Truecharts. 😊 After TrueNAS 24.04, iX-Systems removed K8s and no longer uses Truecharts; they now use Docker/Docker Compose. I can't remember whether Truecharts was the community-maintained version or the TrueNAS-supported version, because there was a community-maintained version that stayed more up to date with the latest application releases. The UI had a feature that let you define a chart URL back then—I miss that. But I think you're right that we're not actually behind. Because 2.2.3 is the latest release of Immich, and that is what is in TrueNAS now. I will say that it does seem to take some time (several months) for them to update their Train with the newest application versions. My Immich Mobile app stopped syncing because it kept updating, and my server version stayed the same. Regardless, I cannot update Immich from the HexOS UI: When I click "Upgrade," it spins endlessly, and eventually it times out: So I went to the TrueNAS UI: Both methods kill the container, and I get this message in TrueNAS: I don't know if anyone else is experiencing this type of non-progression issue. Seems like something got corrupted —could just be me, but just in case, I thought I'd bring it to the thread's attention. Last night I migrated Immich to another server using the REST API and a Vibe-coded NodeJS script.

So I had the same issue and was able to get it to show remote streaming is good to go but my server is showing up as remote even on the same network. This is my first time working with docker and ive been trying to get it to work for days. any advice?

Unfortunately there are 2 version numbers for the apps in the truenas app catalog. The "version" which is what you see in hexos, this is just the iteration of the truenas docker app which has nothing to do with the overall application which is also called "app version" you can see both of these numbers in the truenas interface if you would like but here is a screen shot My "version" is 1.10.12 but that is the same thing as "app version" 2.2.3 so we are not actually behind. i totally get the confusion though and ill be sure to bring it up with the team. Just to clarify a few things Truecharts is a third party app store that afaik shut down and was never compatible with hexos All the apps you see come from the Truenas apps market which i believe is first party to truenas HexOS unfortunately does not have any control over app versions (as much as i wish we did for troubleshooting), this is all truenas

honestly we don't much about when its coming just that it has started development. it was probably more about setting proper expectations that its not gonna come anytime soon. probably best not to build off of beyondRAID since it was slow. slow is bad in so many ways. resilvering is longer, giving drives a longer window where failure would be catastrophic expansion and scrubbing would also take forever. plus don't we all want our systems to be as fast as they could be just a guess but i would assume it has to do with the parity calculations and making sure the zfs stuff works correctly. the forums are here for all your questions so don't be sorry. local interface is definitely still coming. I just don't have an eta to give you.

For anyone unfamiliar, see Docs: ZFS AnyRAID What are the complexities of implementing the AnyRAID feature? I saw a few mod posts saying its still likely years away. The equivalent BeyondRAID on DROBO was one of my favorite features. I'm sure its proprietary and nobody is going to just give you their code. But I'm curious if its possible to build off something like BeyondRAID now that DROBO is defunct? Sorry for being a pain but 2nd to a local interface this is really the main feature that I really want from HexOS.

No, I haven't figured it out. I think HexOS is tightly controlling the Truecharts versions for compatibility reasons or something. I got a notification in the HexOS UI that I can update Immich from 1.7.19 to 1.10.12, but Immich is currently on version 2.2.3. So these versions are waaaay behind. When I look at the version for a new Immich install via HexOS it's 1.10.12, so I don't think this is some upgrade path they're taking—like when you have to go from 1.7.19 to 1.10.12 to 2.0.2 to 2.2.3 instead of 1.7.19 to 2.2.3.

Yup it was added last night! Check the original post for instructions.

Thanks for this, it worked for me too. Before trying this, I tryed to create a new instance manually in truenas but ran into certificate issue.

Nice to see work progressing. Is there an install ISO yet for Fangtooth?

Can guides like this be converted to custom install scripts or it that just to much work?

I just installed HexOS and I'm now trying to install applications, but having issues with every one I want to install. Failed to install app "jellyfin". Failed to install app: Job 205 failed: [ENOENT] Unable to locate 'jellyfin' at '/mnt/.ix-apps/truenas_catalog/trains/community/jellyfin' Failed to install app: Job 155 failed: [ENOENT] Unable to locate 'immich' at '/mnt/.ix-apps/truenas_catalog/trains/community/immich'

Hey, It seams to be possible to use a GPU with multiple apps, but you need to login into truenas to do this: https://www.truenas.com/community/threads/can-multiple-apps-utilize-the-same-gpu.99757/#post-712876

Just wondering how easy/hard it is too time share a GPU between various applications. If I have a single GPU, can it be shared between multiple application tasks. Like Immich ML, Plex transcoding, Jellyfin transcoding, an AI container, etc. I know that you can pass through a GPU to a virtual machine. But not sure if this same mechanism is in play for containers. I guess my ultimate question is if multiple GPUs are required to do justice to multiple workloads across multiple apps. Or do we need to move towards a generic GPU container which the various apps can call upon for their various workloads. Cheers

I had a similar problem with Permissions when setting Custom Host Paths with handbrake on 25.10.0 - Goldeye. I had to set the Owner and Group to "App" for my Folder from the Shell for the Handbrake App.

Thanks for reporting these issues. We are taking this feedback to heart and will work to address.

Seconded. As well as a "Check for Updates" section. It doesn't actually need a "Check for Updates" button, just something that says you're up-to-date and an "Update" button if you aren't up-to-date. Only being able to update from the notifications makes me feel viscerally uncomfortable. I got the same "Update failed" notification, and the "Click to update" notification was gone. I don't like that. I thought notifications were supposed to persist in the archive/history...

I'm honestly not too sure, i haven't tried that yet. I imagine it supposed to see the existing config files and work.

So keen for this, ATM it's taking a minute to do a full refresh. and for comparison, TrueNAS Also, do we know if we will be able to disable the remote deck? I already have a VPN

yes. keep in mind that how metadata/special vdevs work is that your index for your storage drives (like the table of contents) get added to it, which can speed up things when yo have say 1000s of folders with 1000s of folders in them. They also store all the files below a certain size (can be changed via truenas interface) Also apps created via the hexos interface work exactly the same if created through the truenas interface the way to check would be to run some shell commands in the truenas interface, you would get an output like this steps to check 1) enter truenas interface 2)enter the systems tab on the bottom left 3) enter shell 4) type "sudo zpool iostat -v <name of pool>" 5) itll ask for a password, your keystrokes don't show as a security feature 5) add a bunch of small files to your pool 6) run the command again and check if the special vdev increased in size.

I just got a A310 for my NAS. Hoping it will work with Jelly Fin

Great. When it comes to apps, can I just reinstall the curated apps from HexOS and they should restore my configurations?

Just to verify, if I create the pool and vdev in truenas. Even though HexOS sees the pool of 4 drives and shows the 2 optane drives as unused/unallocated (truenas shows correctly assigned) the apps and data installed through hexos like immich would correctly be using the optane drives for metadata? I'm honestly not even sure how to verify, I had set it up and threw a couple hundred photos but couldn't tell (mostly because beyond storage capacity I don't know what i'm looking for to validate) whether or not the metadata was going on the optane or just written to the discs in the pool.

I also had the same error and can confirm I am updated to 25.04 Fangtooth from TrueNAS: Might be a good idea to add a panel like this to the deck to confirm the version number, or if there is already one I cannot seem to find it.