aglyons

Hey @febreeze Thanks for checking out my project. Sorry for the delay getting back to you here. I don't check this too often. I built-in a bug/feature report tool into the UI. The bubble at the bottom right will prompt you to log into my Bug tracker using either Google or GitHub. You'll then see a small form you can drag around the page, enter your comments and also minimize so you can grab a screen shot to paste into the form. It's the fastest, simplest and dependable way to get in touch with me about problems or feature requests. To move stacks into a folder you have to unlock the UI with the Unraid padlock. Then you can right click on a stack and you'll see the move to folder options. The drag and drop is for a feature ice been planning for a bit but isn't released yet. The IP and port details are present in the grid. Not sure why you could not see them. Regarding the icons, you can go through one by one and no pick a new one by tapping the icon image space. But that could take some time. So I've put in a patch that will re-find a matching icon for the container when the current icon path is missing. I've also added in backing up the icons to the flash with the compose files. That should fix things up. Keep the feedback coming.

@JesterEE Couple of questions. try 00.05.02 which is just getting pushed out. - which browser do you use: Firefox or Chrome - How do you open Unraid with http:// or https:// I just discovered myself that Firefox, if you are not connecting to an SSL site it injects significant lag and browser freeze-ups and unresponsiveness. It was a pain in the ass to try and troubleshoot and figure out, but once I discovered that that is the case, when you flip over to an SSL connection, it's night and day difference, smooth, fast, responsive, no lag, So if you're using Firefox, can you try out Chrome? And also try the new release that's just getting pushed out. Other issues addressed as well from some other users.

@JesterEE , any feedback on the dashboard widget? I was a folderview user in the past as well, although I never really used the dashboard part that much. I hope what I built out works for your use case.

00.50.00 is out! Do let me know what you think. Thx, A

Hey @JesterEE, not sure if you saw the edit I made to the primary post about v05.00 release. There are a lot of your 'wants' in there. Including SSO on the support boards. I also built the submission process directly into the UI, so you don't even have to leave the Unraid/StaXX UI. Oh, and a new Dashboard widget. I hope it fits the bill.

@JesterEE I wanted to touch on the features you asked for. Dashboard docker widget integration (Similar to FolderView3) I can look into this. Can you describe what you would want to see there? I think shoving all your services would get the dashboard cluttered pretty quick. Pinning specific items might be useful. But could operational statistics, resource graphs stuff like that, is that something that would be a good addition? Stack/Folder custom icons (Similar to FolderView3) The main stack icon, as it stands right now, is a tiled view of the defined services. If you set the service icons as you like, the stack icon will reflect a tiled composite of them. For folders, I hadn't considered changing the folder icon itself, but I can add that to the possibilities list. Notification/Automerge of an upstream docker compose environment (YAML, env files, etc.) modifications One of my biggest gripes with the current Unraid docker template system is that when a docker template is updated by the author, there is no way for a current user to easily merge the changes into the template deployed in Unraid. Staxx offers and infrastructure that could do this better while not compromising user modifications to the upstream configs. So StaXX does have automatic metadata and project example scanning along with automatic service icon pulling. It's by no means complete, as there's no defined standard. What it does is scan GitHub repos linked to the service, pull in the example compose file, and suggest items your current YAML is missing or items your YAML has that the examples don't. This covers new features that a repo might add to a project that require ENV vars or other accommodations in order to take advantage of. Another function is tracking when an image moves from one repository to another, such as when the source image used to be on Docker Hub but is now hosted under GHIO. It ensures it is the same author and doesn't just point you to some random image with the same name.

Hey @JesterEE , thanks for checking out the project. I hope this is all still relevant once UR8 is released, with them jumping into the compose situation as well. I addressed your concerns, all except one as I can't seem to replicate the problem. Folder Collapse/Expand arrows. They change on my system, and I've reviewed the code there as well. SO I can't explain why you're seeing that. In the future, submitting bugs or feature requests is best done on my board: https://staxxfb.quadcom.ca/. You can paste screenshots directly into posts, which seems difficult, if not impossible, on these boards. I'll be working on adding in-app bug reporting soon that will make the process even simpler. Again, thanks for checking Staxx out. I also really appreciate the feedback and the review!

So the plugin has been on the Unraid CA for a little while now, and I see that there have been some people downloading it, so that's always cool to see. https://ca.unraid.net/apps/staxx-0626nzk0azbrks I would love to hear feedback from people: what they love, what they hate, and what they'd like to see different. Feedback is always great to hear. While you can write stuff in this thread. I'll check it from time to time. If there is something very pressing, then by all means post a card up on my bug tracker/feature tracker that's dedicated to Staxx. https://staxxfb.quadcom.ca/ I really hope this is a plugin that people find useful and really come to like.

CA Submission is in! Fingers crossed ;)

I've been working on this project for the past couple of weeks, and I think I'm at the point where other people can start trying it out! Unraid Community AppsStaXX for UnraidStaXX is in beta: the core is settled and in daily use, but expect the occasional rough edge, and read the changelog before pressing update. Find StaXX for Unraid OS and install it from your server wi StaXXDocker Compose stacks for Unraid, with a form instead of YAML. StaXX is a plugin for Unraid 7.2+ that manages your containers using standard Docker Compose stacks instead of Unraid's XML templates. Paste in a compose file from anywhere, and it just runs. If you would rather not touch YAML, StaXX reads the file and builds a settings form from it, so you can configure a container the same way you do today. Whatever you save is still a plain compose file that works unmodified anywhere else. I've been building in lots and lots of features with a lot of automation as well that automatically fills things in like file paths and will automatically connect the same values that are used in a multi-service stack or even across stacks! Honestly, I can't keep up with all the things that I've been shoving into this! I'm working on updating all of the documentation as well, but you can try it out and let me know what you think. I think it's pretty intuitive. Key Features Runs standard compose files, unchanged Intelligently turns a compose file into a settings form Imports and converts your existing Unraid template containers Imports a Docker Hub image as a starting stack; just give it the URL Groups stacks into folders on one page (Replaces FolderView) Start, stop, restart, recreate, realtime logs, shell and file manager for each service Checks for image updates on an intelligent learning schedule Roll back to a previous image build or historical YAML version independently Pin an image to one exact build Keeps an edit history of each compose file Autostart with the server and chosen order Export a stack as a shareable bundle* Optionally integrates deeply within the Unraid UI CA: https://ca.unraid.net/apps/staxx-0626nzk0azbrksGitHub: https://github.com/quadcom/StaxxFeedback and feature requests: https://staxxfb.quadcom.ca Bug reports and ideas are welcome on the feedback board. Tell me what breaks. UPDATE: 10-02-26 I've had my head down working on the 05.00 release, and this is a big one. Here's a preview of the new features I've added, along with numerous tweaks and bug fixes. I apologize if some of this reads a little weird. Yes, I'm using a lot of AI for this. It is helping me tremendously, especially when it comes to writing out documentation. Writing documentation, coming up with features, and doing all the Q&A is incredibly taxing. So, a little patience with what's written in the help files and these features that are below. StaXX 00.05.00 is almost hereMerge stacks: turn a pile of apps into one stack Lots of apps arrive from Community Applications as separate pieces: the app, its database, maybe a cache. Merge joins them into one stack, and it walks you through every step so nothing changes behind your back. Pick two or more stacks from the Add menu. Each one shows up as a card with its icon, image, ports and live CPU and memory. Name the new stack and choose its folder and icon. Read each original file side by side with the merged one. Every changed line is marked. Hover it to see why, then approve or decline it. You can't move on until every change has an answer. The .env settings files are joined too. Any extra files that clash are sorted out on a folder tree that shows where each file came from. Set start order, health checks, updates and notifications for the new stack in one place. Docker checks the merged file before anything is written. You'll see "Docker validated", or Docker's own message if something is wrong. Your data comes with it. Storage that Docker manages carries over by name, so a database keeps its data without copying a thing. Port clashes are found and settled, even a single port hiding inside another app's port range. Published ports stay open unless you say otherwise, and StaXX tells you if another stack connects to one. Apps that need to talk to each other are put on a shared network, and their addresses are rewired to point at each other. An app that runs through a VPN container keeps running through it. Your comments, notes and shared settings blocks travel with the files. A main file plus an override file is merged as the pair it is. Nothing is deleted. The original stacks are retired: greyed out and kept on disk, with a one-press Remove once you've checked the new stack works. It's built to be careful. Two merges at the same moment can't both go through, and a stack that depends on a file the wizard can't see is refused by name, with the fix spelled out. The user guide has a full worked example: joining the three Tube Archivist apps into one stack, with a picture of every step. Report a bug or ask for a feature from inside StaXX Most apps send you off to GitHub or a forum thread when something goes wrong. StaXX has a feedback form built right in, on every screen. Press the speech-bubble button in the bottom right corner of any StaXX screen, the editor and Settings included. Choose Bug report, Feature request or Improvement. Each one lands in its own section of the StaXX feedback board. Sign in once with your Google or GitHub account. There's no new account or password to make, and StaXX remembers you after that. You can disconnect any time in Settings, Integrations. Give it a title, say what happened, and paste screenshots straight into the text with Ctrl+V. The form floats over the page. Drag it out of the way, or roll it up to the top of the screen while you take a screenshot. Each report includes your StaXX and Unraid versions and the screen you were on, so I don't have to ask. Nothing else about your server is sent. For bugs, you can attach the things that help me find the cause: the stack's compose file, what happened the last time it started or updated, recent container logs, page errors, the self-test, and a record of your last few imports. Passwords, keys, email addresses, network addresses and computer names are swapped out before anything leaves your server. You preview every file first and can hide anything else you want kept private. If a file holds a number that might be an address, StaXX marks it "Check this file" and waits for you to look before sending. Attached files are private to you and the StaXX team, and they're deleted after 90 days. Once it's sent, you see exactly what went out and get a link to your report on the board, where you can follow it. When an import fails, a Report this problem button opens a bug report with the details already attached. Notifications that tell you what happened Update messages now show the old and new version, the size on disk, the first lines of the release notes, and a mark when the major version jumps. A failed update says why in plain words, like a download limit or a tag that no longer exists. New alerts when an app keeps restarting, fails its health check, or stops by itself. An alert if Docker itself stops answering for five minutes, and another when it's back. For each kind of message, choose straight away, in a daily or weekly summary, or off. No summary is sent on a day with nothing to report. Quiet hours hold messages back. Failures always come through. An optional StaXX email with app icons and colours, sent to the address in Unraid's notification settings. Phone alerts stay short: a title and one line. A weekly reminder of anything still pinned, so a pin never gets forgotten. Send a test message and see the result right away. Import from other compose tools Coming from Dockge, Portainer, Arcane, Dockhand or Komodo? Import finds their stacks and copies them into StaXX, even files a tool keeps inside its own container. Have stacks with no containers yet? Point Look in a folder at them, or browse to the folder. More Unraid templates import cleanly now, including ones with settings named only by a number and ones that set the same variable twice. Help fixing a broken compose file A file Docker refuses is still saved. Its row turns soft red with a wrench, and one click takes you to the problem. A floating help window shows Docker's message, what it means in plain English, how to fix it, and a link to Docker's docs. The bad line is marked in red. For one common mistake, Fix it for me shows the change, makes it when you agree, and Undo takes it back. When StaXX meets a Docker error it can't explain yet, it can send it in with names, paths and addresses removed, so an explanation gets written. New explanations arrive once a day. You can switch this off. A StaXX tile on Unraid's Dashboard Arrange your stacks on a grid of up to 8 columns and group them into folders. Give a folder an icon from four icon sets, a built-in symbol, your own picture, or one that cycles through the stacks inside. Right-click to start, stop or restart. Click a stack for live CPU, memory and network graphs. Folders on the Stacks page can have icons too. More control over updates Pin a service to the exact build it's running. Pinned services are skipped by update checks and bulk updates. To unpin, pick a tag from a list. Set update and notification choices per container from the editor or the right-click menu, or for many stacks at once with Select. Recreate on the row menus rebuilds containers from the file. The restart-pending chip can rebuild only the services that changed. Roll back works again, including for linuxserver apps like Plex. Check for updates counts through your images as it goes. Storage clean-up Scan stored images shows a bar of how Docker's space is used: in use, kept for roll-backs, clutter and free. The list shows only real clutter, and the sizes count what removing it actually frees. Clear out what Unraid Docker and Compose Manager left behind, like old templates. StaXX keeps a copy you can put back. Damaged containers are flagged and can be cleared. If Docker won't let go of one, StaXX offers to restart Docker and try again, and warns you first. No more scheduled image removal. A notice tells you when clutter builds up, and you decide what goes. Removing a stack also clears roll-back images that nothing else needs. Proxy and DNS built in Connect Nginx Proxy Manager and Pi-hole in Settings, Integrations. Add a domain, certificate, WebSockets and a Pi-hole entry to any app from the editor. StaXX shows what it will change, then does it when you press Apply. A small DNS mark on the stack list turns green when everything matches and red when it has drifted. Click it to fix. A clearer stack list Status chips follow one colour scheme: green is working, grey is resting, blue is good to know, amber needs you, red is broken. A legend explains every mark. One Add button holds New stack, Apps, Import, New folder and Merge. Pick stacks with Select, and Start, Stop, Restart, Check and Update appear right below it. CPU now shows a share of the whole processor, from 0 to 100%, and each graph has its own colour. Apps with a health check get a gently breathing green chip. Apps that update themselves get a small green mark. StaXX tells you when it can't see the ports of apps on the server's own network. If one can't start because its port is taken, it names the port and what's using it. Editor improvements Drop a picture or a picture link onto a service's icon to change it instantly. Each service keeps its own icon, and a stack shows them tiled together. Fold away any section of the editor. StaXX remembers what you folded. The Image box suggests tags: the usual choices first, then version numbers and images already on your server. Click a setting that uses an .env variable and its line lights up in the .env file. Download a container's whole log as a file from the Manage tab. The file manager has its own on/off switch. Settings A tidier layout, with the self-test on its own tab. It runs when you open it and has a Copy button, ready to paste into a post here. A new About tab shows your StaXX version and which Docker Compose is running. The backup check now covers your whole StaXX folder, settings included. The copy kept on the flash drive is rewritten only when something really changed, so the flash gets written to less. A rewritten user guide The guide now reads as a step-by-step manual, with close-ups of the controls each page points at. 00.05.00 is in final testing and should land very soon. If you try it and something's off, the new feedback button is the fastest way to let me know. Thanks to everyone who has been testing and sending in ideas.

I was on the ":latest" tag forever, along with the "latest" env var, so it was always bleeding-edge, not a beta version just the very latest releases. I tried a few different tags before I hit this one, from I believe 8 months ago, that started working properly again. During this time, there were no definitive logs to show what was happening. All that would happen is the GPU would just stop handling the transcode, and it would silently flip to the CPU. I wouldn't even notice anything different about the stream. But the CPU would jump to 100% for a single stream. I should add that it also ONLY ever happened when tonemapping was involved. A straight 4k SDR to 1080p wouldn't crash at all.

***UPDATE: Downgrading Plex to tag 'plex:1.43.1.10576-06378bdcd-ls300' has resolved the tonemapping via OpenCL so all is back to working again on i915 with the A380. Tested with 5 4K HDR tonemapped 1080p streams. Flawless with headroom. They've broken the pipeline somewhere since then Careful you don't get bit by the built-in update catch! If you have an ENV var set to "latest", Plex will update itself inside the container, so it won't matter what image you point at. Set the env var to "docker"; that will lock the Plex version in the container to the version specified by the image tag. I'm running the Linuxserver image, so the full image is lscr.io/linuxserver/plex:1.43.1.10576-06378bdcd-ls300

Is this not the support forum for the Intel GPU TOP plugin? The plugin support link sent me here. Fairly recently I've been having problems with my A380 with Plex where HW transcoding starts but after a few minutes it falls back to CPU. I've tried everything so far that I can find, blocking the i915 and forcing XE. But XE doesn't support tonemapping so it's back to i915. I've now started down grading Plex releases but havn't found one that's stable in this regard. I have posted a thread in the general support but I thought I would come here for more specifics. ***UPDATE: Downgrading Plex to tag 'plex:1.43.1.10576-06378bdcd-ls300' has resolved the tonemapping via OpenCL so all is back to working again on i915 with the A380. Tested with 5 4K HDR tonemapped 1080p streams. Flawless with headroom. I'm running the Linuxserver image, so the full image is lscr.io/linuxserver/plex:1.43.1.10576-06378bdcd-ls300

I figured I'd ask here first in case anyone else has run into this problem and found a solution. Let's compare notes; I'm running a smaller AMD 8500G on an Asus B650M board. A380 is in slot two, limited to 1 PCIe lane, unfortunately. The disk controller is in slot one for the lanes.

Very recently I've started to notice that any 4k HDR source that transcodes (and tonemaps because, yes it needs to) will randomly crash the encoder and fallback to CPU. This came up trying to troubleshoot another issue in that I have an AMD iGPU that was caught in the dri mapping snafu. You thought you were using the Intel when in fact you were running the AMD. Result=glitches, corruption just unwatchable. I've been going down a rabbit hole trying to get HDR tonemapping while transcoding workg again and every approach I go down has been a failure. I now come to the understanding the using the XE driver can't do tonemapping at all. It can transcode down in scale but can't tonemap. If that's on, it straight to the CPU without even trying. What does Unraid need to have in order to get this working?! And I am floored that I don't see anyone else talking about this, unless I'm the only one that has something borked!

Was the reason for moving to the XE due to performance? I just pulled my older Nvidia card and put in an A380, and my first test was a major disappointment. One 4KHDR>1080 stream, and the GPU was hitting 94% usage. No better than the Nvidia card I had just pulled. Does the XE driver fix this? Also, is the ":plexpass" tag for lscr.io images?

My intention was not specifically JUST Unraid, but also the containers and VM's running on the system. I ran into an issue with Penpot using compose, where the exporter container was crashing and restarting so quickly that the UR UI couldn't report the restart loop. Only looking at the log files informed me of what it was doing. If this internal AI had been watching the logs, it could have informed me much earlier. As it stands right now, unless you are watching logs on a regular basis, which I suspect the majority of UR users are not, there could be issues happening in the background, causing other issues that are hard to trace for the average user.

I came across an AI log file scanner that would diagnose issues and humanize responses by explaining what is going wrong. It also provided steps to resolve the problem. I commented on the GitHub project, suggesting that a local AI agent could monitor Unraid/Docker/VM logs in real time and alert the sysop to any issues it finds, along with relevant solutions. Think "Fix Common Problems" on steroids. I did not receive a response from the maintainer. GitHubGitHub - YEDASAVG/Stratum: AI-powered Log Intelligence Sy...AI-powered Log Intelligence System - Semantic search, anomaly detection, and root cause analysis for logs using RAG techniques. Built with Rust, Axum, NATS, ClickHouse, and Qdrant. - YEDASAVG/Stratum

Currently, on the Docker page, whenever a container starts, stops, updates, etc., the entire page refreshes to display the container's current state. Using the compose plugin, containers are orphaned when the stack is updated. Clearing out orphaned containers can only be done one by one and requires a full page refresh for each. This takes more time and places more demand on system resources. An ASYNC refresh of changed content would dramatically speed up the process and enhance user experience. Any area of the UI could employ the same ASYNC refresh backend, further enhancing the user experience and system responsiveness.

I suggest allowing only approved individuals access to download the diagnostic files that community members upload. I understand that the data is sanitized and redacted. However, whenever I upload a diagnostic file, my UniFi IPS goes off the charts with blocking requests to my server. I get that this is a community and everyone is here to help each other out. But there is no way to be entirely sure that everyone here is being honourable and decent. I can only speak as to my experience and what I see happening. Perhaps only allow those with a certain number of verified answers to other posts? Forum admins? Forum moderators? Could there be further obfuscation of the data contained in the diagnostic files, such as any domain names or internal/external IP addresses? My network was relatively quiet regarding penetration attempts, but I recently posted a thread and attached my diagnostic file. Ever since then, I have been logging over 100 attempts a day to various ports and to my NPM container. And I'm a home user, not a big corporation. Thank god I went with the Unifi gateway to help block all this. Not cool. A.

I'll run the iperf. Just so I am clear, if an array folder has a cache-first config, when writing to the array, the files will initially be copied onto the cache pool, and then the mover 'moves' the files to the array. If this is not the 'understood' function, do I have to specifically state the files are landing on a cache pool, regardless of the fact that the mover will relocate the files? I chose to be specific about the config in the event that a pool array behaves differently than a shared folder cache config. Yes, they both land on a pool device, but in the latter, the mover is involved, which may introduce other parameters.

I did state that the files are being transferred to an SSD ZFS mirror cache drive. 2025-07-21_10-02-35.mkv 2025-07-21_10-02-04.mkv

I've posted about disk throughput before, and I have seen a number of other posts about it as well. So, I'm back with another question, hoping to either come to an understanding of why it works this way or potentially expose an underlying issue. I have two Samsung 870 EVO drives in a ZFS mirror as a cache drive for a shared folder. Everything written to the share goes to the cache (obviously). My network backbone is 10G, although the UR server utilizes the MB's built-in 2.5G NIC. I was transferring a folder with a series of +12GB files over to the array, and the network transfer rate was about 30-40MB/s. Which is a far cry from 10G speeds. That's when I took a look at the UR dashboard to see transfer rates and disk throughput. The network throughput was showing approximately 600 Mb/s, and disk throughput was fluctuating between 0 and 170 MB/s. There was never a constant write speed to the cache drive for the shared folder. The mover is not running, and no other transfers were running in the background. If there are any tests I can run, please let me know. Diag attached. Thx, A.

I think what you're really looking for is not really multiple arrays, but a cache pool for the existing array. None of the arrays (other than a ZFS AFAIK) have any striping, so you wouldn't see any real performance benefits with an array of NVME drives. I think you might even see a performance degradation (compared to single mounted NVME) as the array functions require all disk activity when writing to a disk in order to calculate the parity bits. That is the reason Unraid arrays don't have spectacular throughput performance and the addition of cache pools to speed up writing and reading (in certain configs).

I don't think disabling IPV6 or anything else does it. I suspect that once UR is up, making any change in the config resets the networking, and the SHIM gets created as it should. I myself had the default bridge disabled, and when I re-enabled bridging on the ETH0, everything started working. But I have seen many other people saying they enabled this or disabled that, and it started working again. I really think the problem is rooted in the boot-up process and the creation of the network at boot-up. Perhaps the SHIM creation process is run before another function that is required for the SHIM, in which case, the SHIM creation would fail. But once the system is up, changes to the networking runs the SHIM process again, and because all the other functions are already in place, the SHIM is created correctly. Again, this is all theory. I don't possess the required Linux acumen to diagnose this fully.