• 0 Posts
  • 16 Comments
Joined 1 year ago
cake
Cake day: June 13th, 2023

help-circle
  • Yes, in the sense that you are responsible to update the Docker container and often this can lead to vulnerable containers. No, in the sense that it is much easier to scan for dependencies inside a Docker container and identify vulnerabilities. Also most containers are based on Linux distribution, so those distribute the security fixes for specific libraries. All you have to is update the base image.


  • julianwgs@discuss.tchncs.detoSelfhosted@lemmy.worldAlternative to Minio ?
    link
    fedilink
    English
    arrow-up
    2
    arrow-down
    1
    ·
    1 month ago

    Dependency-free doesnt mean they dont have dependencies. Its just that they bundle them all in the executable. When there is a security vulnerability in a library on your Linux system the vendor of your distribution (Canonical, Redhat, SUSE) takes care that it is fixed. All dependent software and libraries are then fixed as well. All I say? Not the ones which have been bundled in the executable. First they need to find out that you are affected and then the maintainer has to update the dependency manually. Often they can only do this after there has been a coordinated release of the fix by the major distributors, which can leave you vulnerable no matter how fast the maintainer is. This is the way it is in Windows. (This was a short summary)



  • Are you just starting out? I got started with home labbing with a Raspberry Pi 2B (1GB RAM!) and an external HDD I had lying around. I host Yarr, Navidrome, backups and a dashboard app Ive written on there and I am quite satisfied. I would really recommend starting small with hardware you already have and then buy new hardware as you go along. I am also using Tailscale. With this you can get your initial setup up and running in a day and save money if it turns out home labbing isnt for you or you dont really need the hardware.






  • julianwgs@discuss.tchncs.detoSelfhosted@lemmy.worldCost-cutting tips?
    link
    fedilink
    English
    arrow-up
    14
    arrow-down
    1
    ·
    9 months ago
    • Use sqlite instead of Postgres, MariaDB
    • Avoid enterprise software (Kubernetes, Elastic Search)
    • Only use projects with efficient programming languages such as Go, Rust, etc.
    • Try to run things bare metal
    • Lookout for projects which name themself minimal or light-weight

    I use a Raspberry Pi 2 to self host a Dashboard written in Rust (Axum), a RSS reader called yarr and a music streaming server Navidrome. The latter two are written in Go and very resource efficient. The electricity bill should be under a Euro a month (6.4W max power consumption).



  • This pattern is true and passenger kilometers represent it just fine. There is no need to use the how often you use the train metric. Note that my two examples were there to explain the metric, not actual factual examples.

    As an actual example: I take my bike to work and dont own a car, so my modal split is mostly trains because of longer distance trips, but I use the bike far more often. Frequencies only make sense if each occurrence is very similar (in quantity). For example: How often does one eat meat? Each meal roughly contains the same amount of meat (may be factor two or three difference). Here frequencies make more sense as more detailed statistics dont actually give more insights.


  • Thanks for your comment. Not wrong in the sense that the data is wrong or faked, but that the metric is not useful. Especially when better metrics are readily available for that region. Can you name me one prediction or result which you can infer from the frequency of train travel other than „fun facts“? (I am actually really curious :) ). With the modal split you can for example calculate CO2 emissions or estimate needed capacity increases if you want to replace one mode with another and much more.



  • Not quite, it‘s only restricting competitors and so all companies and home labbers can still use it for free and contribute as in free speech.

    However this can bring a lot more financial sustainability to a project. I don‘t know the specifics, but the main problem is that companies make profit of the software, but don’t invest enough money back into the product. This cannot be good for users. Open source must be financially stable.

    Also right now all those competitors (and users) can create a fork and maintain it. So it is up to the community what will happen to the project.