On a very specific note, I don’t run my Plex server in a container. I have a docker compose setup with 20+ apps, but Plex is on the bare metal OS because it’s kinda finicky and doesn’t like nas. You also need to setup the Plex API to claim the server as the container name changes. This is my stock Plex config if it helps

plex:
    image: lscr.io/linuxserver/plex:latest
    container_name: plex
    network_mode: host
    environment:
      - PUID=1000
      - PGID=1000
      - TZ=Etc/GMT
      - VERSION=docker
      - PLEX_CLAIM= #optional
    volumes:
      - /home/null/docker/plex/:/config
      - /x:/x
      - /y:/y
      - /z:/z
    restart: unless-stopped

It’s built on the shipping container parallel. In order to transport objects you obfuscate anything not required for shipping a container.

• What’s inside the container doesn’t matter. The container has everything it needs to run because the ship/host is responsible for the overhead.
• containers move. Containers are setup to run by themselves, so you can move it from one ship to another. This means you can use your container doesn’t care if it’s in the cloud or a shipping vessel
• As soon as you open a container your stuff is there. It’s very easy to onboard.
• Most importantly though, your shipping container isn’t a full boat by itself. It lives in a sandbox and only borrows the resources it needs like the hosts CPU or the boats ability to float. This makes it easier to manage and stack because it’s more flexible

Proton had a reputation for being the good guy. In the span of a month, we saw them bend the knee, flip flop and throw shade at competition; all while pretending to be the hero. We essentially have to trust them with our data and they are showing signs that they are willing to act against that trust with worrisome agendas and biases. It’s not a good look, and since this marketing to users key issues, it’s going to cause some responses.

This thread is a dumpster fire, routing infrastructure, solar panel addresses, we are adding this to EVERYTHING WE ALREADY HAVE that is growing exponentially. I work on an L7 support team, regular users are clueless on how this stuff is setup and apparently have strong stupid opinions. Anyone still reading disable ipv4 in your home network and try to roll forward. You will fail, and finite numbers are finite.

So, yes a few pieces of land mass tech such as smart road or solar paneling and we hit the theoretical limit of IPv6. And we currently dont need the addresses. So glad that you agree

I’m sorry but how? We have appliances with dockerfiles, micro containers for remote controls, extensive botnets of virtual machines, centuries in the future when we have expanded into the solar system and trillions of humans all having millions of unique applications with addresses, it’s inevitable to hit a finite number. When every square meter of smart road has an routable address; we will likely be rewriting networking anyways. The only players pushing IPv6 transition are networking companies because a new standard requires new hardware.

Cisco as a client tried to force ipv6 for their managed service and after an entire quarter of attempting to resolve it, we actually disabled it for their virtual address per their request. IPv4 has issues and IPv6 promises solutions, but it’s not a stable platform yet. This appears ignorant but is based on truth. IPv6 is also eventually going to hit exhaustion with the frequency we spin up virtual machines, it’s okay to skip a bad generation.

It’s released but it’s insane feature light, has massive injestion problems, requires massive collection overhead and doesn’t have a fraction of splunks indexing. And it’s using the standard dd UI and I personally dont like. Logs aren’t metrics, they need a different interface.

Datadog logs are basically in beta. You can send them synthetics apm and rum but I would be interested in spinning up my own private greylog instance to get away from DD logs

Skimmed most of the thread and there are a lot of guesses, the actual answer is presumably impossible given the parameters. Asset management tracking software is pretty much permanent tracking these days, screen idle time, keystrokes per minute, application focus tracking. A lot of higher end devices have gps chips in them by default, your works VPN reports it’s trace route so it will have general geographic location. Microsoft and Google cooperate accounts even offer remote hard drive wipes to protect company secrets, regardless of the location. My work gets reports of where people connect from as part of the RTO policy. We had someone working from their parents house for a few weeks and got emailed by HR asking why they were logging in from an unapproved area. Most places can pull this data, but not all of them act on it.

Network engineering is kind of in the middle where you take the skill set of help desk and office management. This often leads to help desk and software development both falling under the organization in information technology. Application support also often falls under this category.

I use this in my bash profile in case anyone else finds it helpful. Usage is:

gao fixing a typo

function gao() {
    git add .
    git commit -a -m "$*"
    git push origin `git rev-parse --abbrev-ref HEAD`
}

I’m going to be late to the next meeting, I can’t quit vim ☺️

You must have struck a nerve or that user has other stuff going on. Several months ago AI was practically eating paste. It definitely has short comings now but the adaptive speed is definitely going to disrupt most markets. AI will be considering things that the users aren’t aware of and have significantly more training than humans faster.

“I wish I was 127.0.0.1” sounds perfect for the situation

~“Select shirt from closet where clean == true”

“Select finger from hand where index=3” if you are cheeky

“+1 cold resistance, +1 style, +1 armor” if it’s a gaming group

“418” with a picture of a teapot, fringe but it’s the unique http error code

It’s because Elsa was originally written as a villain until Let It Go was written and they pivoted the story so it’s less one dimensional than most Disney films

I started naming all of my devices based on their number mk1-mk6, then I realized it’s hard to differentiate desktop from phone off a list. Now I’ve got: Mkr1 - rack server Mkp6 - phone 6 And desktops/laptops are just mkX depending on age. Then set static IPs based on their number