• IsoKiero
    link
    fedilink
    English
    arrow-up
    4
    ·
    9 months ago

    Why billboard system would have sane installed? I don’t think Debian or derivatives install it by default. Vnstat is also a bit odd, but maybe that’s just me. I assume they have multiple of these displays around and for them it would make more sense to use something more centralized, like zabbix, to monitor the whole network (obviously they could do that too).

    • corsicanguppy@lemmy.ca
      link
      fedilink
      arrow-up
      5
      ·
      9 months ago

      assume they have multiple of these displays around and for them it would make more sense to use something more centralized, like zabbix

      The one I saw a decade ago yielded SNMP to solarwinds (I know I know) rather well, but they mainly used PING on it to see when the radio link died.

      Fancy that – when the parks n rec sites were converted to e-billboards, they had power but no net line, and “radio’s fine”. Show me an old linux billboard host and I’ll show you a canvas my inner child can’t wait to e-graffiti.

      • IsoKiero
        link
        fedilink
        English
        arrow-up
        2
        ·
        9 months ago

        Wait a second. They used AMPRNet to manage these things? In here this kind of things are either hardwired to the internet or they use 3/4/5G uplink and while of course techinally possible either way to breach the system it’s a bit more difficult to find out proper IP’s and everything.

        Once upon a time I had a task to plan a scalable system to display stuff on billboards and even replace printed ads on stores with monitors. The whole thing fell down as we couldn’t secure a funding for it, but I made a POC setup where individual displays had a linux host running and managing the display with (if memory serves) plain X.org session with mplayer (or something similar, it was about 20 years ago) running on full screen and a torrent network to deliver new content to them with a web-based frontend to manage what’s shown on which site. Back then it would’ve been stupidly expensive to have the hardware and bandwidth on a single point to service potentially few thousand clients, so distributing the load was the sensible solution. I think that even today it would be a neat solution for the task, but no one has put up the money to actually make it happen.