It’s been a while! Let us know what you’re currently working on, what exploded, what cool stuff you found, or just brag that everything is running fine unattended.

I haven’t been doing a lot, renovate auto upgrades for patches and sometimes minor updates work great so far. I got involved in my union to spin up a mastodon server soon, so there’s a new project, but I already run mastodon so it shouldn’t be much of a deal, I just need to get comfortable with the infrastructure. I also set up ntfy but only sent some test notifications and didn’t configure any app yet - planned for Element Chat.

[Meta] I’m not sure how to apply the new tags. In the recent versions of this thread, people sometimes posted about their ollama setup or the like, but I’m not mentioning AI in the post body (except this meta thingy) and it shouldn’t be much of the discussion, but it’s an open post. Maybe curbstickle can chime in to let me know if I should apply one of them.

  • InnocentZero@kbin.earth
    1·
    14 hours ago

    https://lemmy.world/post/49663813

    My old post from my old account.

    Since then, I’ve been moving my PC (not server) from arch to nix, and discovered that I was relying on fundamentally broken DNS resolution behaviour on netbird.

    Fixed that and also changed my Caddy config to use FQDNs as well.

    Apart from that, added Gollum as a wiki server, and the experience has been perfectly amazing!

    Still waiting to get my own mini-PC and use that, current shitbox is a desktop in my college.

  • SayCyberOnceMore@feddit.ukEnglish
    16·
    2 days ago

    I found the best thing to deal with a device that won’t work…

    I’m trying to get the 3.5" adafruit touchscreen working and it’s being stubborn…

    So… I went outside and split a large pile of logs.

    Problem still not fixed, but I feel a hellofalot better

  • Victoria@lemmy.blahaj.zoneEnglish
    6·
    2 days ago

    Fiddling around with Renovate Bot on my Forgejo instance. It’s not working perfectly yet, but it’s getting there.

    Also I began using LubeLogger to track my car expenses and fuel efficiency. Works beautifully.

  • Carl Newton@feddit.ukEnglish
    20·
    2 days ago

    I built a dashboard themed around my dog. It uses the Home Assistant API to get various stats from the house. I’m pleased with it!

    • undefinedTruth@lemmy.zipEnglish
      3·
      2 days ago

      Care you share any more details? Looks very aesthetically pleasing. I think I should make one themed around my cat.

      • Carl Newton@feddit.ukEnglish
        3·
        2 days ago

        It’s a PHP container On the front end I’m using Bootstrap and theming it with Catppuccin. The data for the home energy charts are populated from the Home Assistant API. They’re drawn with chart.js. The server stats are pulled from PHP exec commands, and drawn with Bootstrap progress bars. The temperature card also takes data from the Home Assistant API, which also has an endpoint for local weather data. This doesn’t update as often as I’d like. But it compares the outdoor temperature with the reported temperature of my thermostat, and then the messages below that are determined by a comparison of those values. I’m also using HTMX to load each card, and each tab of the energy card separately so that I don’t have no wait on all requests for the dashboard to load. Not that it’s slow - the home assistant API is surprisingly efficient at loading a great deal of logs. When you click my dogs face, she says WOOF WOOF WOOF in a modal to offer me a search with searxng.

  • yuman@programming.devEnglish
    4·
    2 days ago

    got a busted Lenovo Ideapad 520ISK out of a dumpster, missing screen but otherwise perfectly functional. the thing has i5-7200u, 8 GB, two SATA ports inside and this prompted me to look up converting the thing into a server, specifically figuring out how to run 3.5" drives off it; I’m very anti-friendly towards using USB solutions for such a task.

    searching was arduous, both web and yotube, and except one lonely forum post (remember those?) didn’t find anyone attempting let alone succeeding in running it thus. the 2.5" and 3.5" connectors are identical (there are micro SATA to SATA adapters for the DVD port) but the drives will need 12 V that the 2.5" ports ain’t got.

    what I’ve concluded is that all that’s needed is a M/F SATA extension cable which I have trouble sourcing locally; the power will come from a separate ATX PSU for the time being, although I’m sure I can pull 12 V from somewhere on the board.

    don’t got no 3d printing nor fabricating options, so it’ll likely look super-janky, but that’s not a huge concern for me now. I’ll do a write-up if I succeed.

  • BakedCatboy@lemmy.mlEnglish
    1·
    1 day ago

    I completed a 300 mile move yesterday, and my homelab was one of the least stressful parts thanks to my pre-emptive downsizing from a half rack to just a Synology + Intel nuc. I also pre-emptively switched from forwarded ports to a reverse proxy because I didn’t want to be thinking about what isp I would end up with. Even further, I linked the nas and nuc directly with Ethernet so that I wouldn’t need to rely on static IPs or router-assisted local hostnames, mdns, avahi, whatever because I wanted to be able to transplant my lab onto any network just in case of emergency - thankfully I haven’t had to test that out but it’s nice knowing my lab should work on any network where both boxes can reach the internet / my reverse proxy. I was thinking worst case if we ended up in some apartment complex with some crappy provided setup that had broken local discovery then it would be 1 less thing to worry about.

  • MuttMutt@lemmy.worldEnglish
    1·
    1 day ago

    I have so many irons in the fire I can never quit…

    Right now I am tinkering with a local only HomeAssistant AI voice assistant. I have an Ubuntu VM running docker on my TrueNAS which has an ARC A310. The reason I’m running a VM is because I’m using A Dual Edge TPU module for FrigateNVR while using the ARC A310 for video decoding. Ollama is running Qwen2.5 3B. I’m also running Wyoming Whisper and open-webui for the voice AI. I’m using a ESP32-S3 AI Smart Speaker Dev Board for voice control.

    I found a project on GitHub that looked really good but was designed to always pull files and so I forked that to create a local only version and setup a small docker container for the audio files. Still working on modifications of the repository but the setup is working OK but I’m trying to work out the bugs.

    I’m also working on a Halloween project. I have a plastic pumpkin with a big red arcade button for a nose, LED’s for eyes, a speaker, some car horns, proximity sensors, and a temper… I call it my Doom Pumpkin. It will go outside a week before Halloween and complain about people touching it’s buttons or lack there of during the day and stay quiet at night. On Halloween it will keep going till later and when someone presses the button it will randomly honk horns in different ways. When someone steps in front it will say funny things to get them to press the button. If they walk behind it will warn them something bad might happen (like saying it farts fireballs.) It’s all setup in ESPHome and will be controlled by HomeAssistant and NodeRed. There is also a theft prevention mode built in that will cause it to just spazz out and blow the horns.

    I’ve still gotta pull another 500 feet or so of ethernet then get those and some others dropped into the server closet and shift some stuff around so it feeds into the rack. I’ve also got to pull my current camera lines from my POE injector and run them into my POE switch.

    This is all on top of building and installing racking for solar, getting the solar installed, pulling a wall in a couple inches at the top where it pulled away during an earthquake before we bought the place, and fixing a bunch of other stuff.

  • f3nyx@lemmy.mlEnglish
    6·
    2 days ago

    spent some time last week adding metadata to my mp3 music obtained with yt-dlp. used musicbrainz Picard.

    I’m actually really surprised how comprehensive that database is, and how easy it was to manually tag the tracks that didn’t get picked up automatically.

      • pienoyer@piefed.socialEnglish
        1·
        1 day ago

        Can only speak for myself. I went with PocketID for oidc. The learning curve was a little steeper than most other homelab services i try, but its 111% worth the setup and time. Logging into services now happens automatically. And if you’re distributing services to friends and family then oidc is a lot more seamless than 10 different accounts on 10 different services.

        I went with PocketID because it looked the simplest out of the bunch. Authentik does provide more customization but afaik this is more needed for enterprise graded stuff.

        All in all, very happy with it, not going back to traditional logins any time soon.

        • Alfredolin@sopuli.xyzEnglish
          1·
          12 hours ago

          Woah, thanks for the tip.

          I just tried pocketid and it’s actually not that complicated. Nice.

  • neidu3@sh.itjust.worksEnglish
    7·
    2 days ago

    Not working on anything particular, but I’m going on a road trip on Tuesday to this regional airport that is giving away two racks perfect for my use case. Deep enough for my larger machines, sound proofed and 42U tall.

  • kokomo@reddit.kokomo.cloudEnglish
    6·
    2 days ago

    Not to brag but everything’s working fine, clean as a whistle. Upgraded forgejo to the latest version. I’m still helping maintain the awesome-selfhosted list, and get inspiration from selfh.st a lot (big shoutout!) Hope to make sure some postgres dbs are up to date at some point, all of my dockhand instances that i set up have been a big help :)

  • brucethemoose@lemmy.worldEnglish
    9·
    2 days ago

    [AIT] I know this isn’t everyone’s cup of tea, but I’m excited about Deepseek V4 flash. It’s (for me) the perfect size and architecture to self-host an LLM.

    My box (and brain) are chugging through a queue:

    • Figure out why my swap is going crazy, and how to ban processes from it [Done].

    • Figure out why Code OSS is unhappy [Partially Done].

    • Make an ik_llama.cpp iMatrix for Deepseek V4 [Done].

    • Figure out why quantization isn’t working [Done].

    • Make a test IQ2_KL/MXFP4_R8 quant to see how it does squeezed onto my box [in progress].

    • Test. Tune. Inevitably troubleshoot the dozen other things that go wrong. Figure out how much spare RAM that leaves me.

    • Make a higher quality IQ3_KT quantization. This will take all night on my CPU.

    • KLD test both of them vs the full precision, to quantify quantization loss. Likely an overnight task, too.

    • Try merging the new model release with the base model, 50/50, for a less “deep fried” model. imatrix, quant, test.

    The goal is to host it on a single RTX 3090, Ryzen 7000 with 128GB RAM, for anyone curious. Though I may try smaller models too, like Laguna S1.

      • brucethemoose@lemmy.worldEnglish
        3·
        2 days ago

        • It’s just under 300B, trained at FP4; absolutely the perfect size for servers with 128GB-192GB CPU RAM to spare.

        • Its fast. I’m getting 17 tokens/sec on a single RTX 3090 GPU, all experts offloaded to RAM; for a 300B model this smart, that’s crazy fast.

        • Its attention mechanism is cutting edge, good for long context without too much processing time.

        • The benchmarks for coding/agenic usage are absolutely bonkers, within margin of error of frontier models or Deepseek Pro in some cases: https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-0731

        • …Though I don’t put much stock in benchmarks. And I haven’t tested it enough to tell you if it lives up to that hype for specific use cases.

        • Deepseek also publishes its base model. That means I can “unfry” the model with a merge if I have to.

        I am afraid of the the model being “overfit” to coding and agenic stuff.

        For reference, my previous favorite model was Xiaomi MiMo V2.5 310B. It benched well, but it also feels “smart” outside of benchmarks, like in knowledge of trivia without tool usage/internet access or comprehension of weird questions.

  • OpenAltFinder@lemmy.worldEnglish
    3·
    2 days ago

    I just published the weekly report for OpenAltFinder: #31

    Added 19 new alternatives, and all projects in the CMS gained a total of 26,373 stars! The top gainer was Pi, with 2,302 new stars.

      • MyRobotShitsBolts@lemmy.worldEnglish
        1·
        6 hours ago

        I think my Postgres docker became unsupported. I think Immich needs Postgres with vector to run and mine did not have vector. I tried to save my data but blew it up in the process. Nothing of great value was lost and I keep all my images elsewhere in my nas for just this reason. I do kinda wish Immich could just look at a directory rather than keeping its own file structure.

        • cojo5@sh.itjust.worksEnglish
          1·
          6 hours ago

          immich supports external libraries, but i’m not sure i’m you get all the functionalities when using them. i just got started with immich so i’m still learning the ropes. actually im currently devating whether i want to migrate my photo archive to immich or just leave it as an external library.