6 ms·

I've recently discovered the joys of vscode-server[0] for this type of use case. Which you can either serve as a WebApp direct from your machine, or tunnel in

I've recently discovered the joys of vscode-server[0] for this type of use case. Which you can either serve as a WebApp direct from your machine, or tunnel in from another VSCode instance - including a webhosted version from Microsoft[1] I may not do that much coding these days, but I already have found it handy for when I forgot to do a Git push! Just log back in to the Web Environment and everything is how I left it. [0] https://code.visualstudio.com/docs/remote/vscode-server https://code.visualstudio.com/docs/remote/vscode-server [1] https://vscode.dev/ https://vscode.dev/

notpushkin 2y ago
There’s also code-server by Coder, which you can just run on your own server and open in browser directly. It’s also open source (most tunneling features in VSCode aren’t). https://github.com/coder/code-server https://github.com/coder/code-server There’s a one-click setup for it in Lunni, a Docker dashboard I’ve been working on (shameless plug): https://lunni.dev/ https://lunni.dev/
- BOOSTERHIDROGEN 2y ago
  As someone who doesn't code professionally, what is the primary use case that leads to significant daily enhancements for developers?
  - lvturner 2y ago
    This is a multifaceted question, because there isn't one use case, it depends on who you are. If you are somewhat nomadic and always moving systems, having a "home" you can dial back to and pick up where you left off from is a massive boon (with a remote system like this, my phone is now powerful enough to do pretty much any development task as now it's just rendering the front end and not doing real heavy lifting) If you're an enterprise programmer working on locked down networks and devices with more end point protection tools than endpoints, an unencumbered authorized system with access to the right network assets, as well as standardized and project specific tooling is the difference between coding and coding with one arm tied behind your back and a blindfold on. (actually the benefits for enterprise are many and not just DevEx -- think security, network admin, vendor access, governance, AI rollout, reduction in project specific on-boarding documentation....) I'm sure there are many other use cases, though these are just the immediate ones that spring to mind.
    - BOOSTERHIDROGEN 2y ago
      I appreciate the commentary; it provided a helpful overview of the use case.
- TheTaytay 2y ago
  Lunni looks cool. I hear mixed things about Docker Swarm. How has it been to work with?
  - notpushkin 2y ago
    Thank you so much! Personally, Docker Swarm works great for me. I was worried about it for a while, too, but Mirantis seems to maintain it pretty well. There are even new features being shipped, like CSI support back in 2023! I’m thinking about adding a Kubernetes backend though, just in case things get worse for Swarm.
- indigodaddy 2y ago
  Lunni looks neat. Is persistence baked in with the docker stuff you deploy on and/or how is persistence handled/achieved?
  - notpushkin 2y ago
    Thank you! We use the standard Docker volumes [1] for persisting data. It’s pretty straightforward – you just define a volume and specify where to mount it: volumes: app-data: {} services: app: image: example.com/app/backend:latest volumes: - app-data:/data This should work pretty well for most use cases. (There are some caveats if you decide to run multiple Lunni servers in a cluster – basically, every node would get its own volume by default – which is the main reason we don’t officially support that yet :-) [1]: https://docs.docker.com/engine/storage/volumes/ https://docs.docker.com/engine/storage/volumes/

notpushkin 2y ago

There’s also code-server by Coder, which you can just run on your own server and open in browser directly. It’s also open source (most tunneling features in VSCode aren’t). https://github.com/coder/code-server https://github.com/coder/code-server There’s a one-click setup for it in Lunni, a Docker dashboard I’ve been working on (shameless plug): https://lunni.dev/ https://lunni.dev/

BOOSTERHIDROGEN 2y ago
As someone who doesn't code professionally, what is the primary use case that leads to significant daily enhancements for developers?
- lvturner 2y ago
  This is a multifaceted question, because there isn't one use case, it depends on who you are. If you are somewhat nomadic and always moving systems, having a "home" you can dial back to and pick up where you left off from is a massive boon (with a remote system like this, my phone is now powerful enough to do pretty much any development task as now it's just rendering the front end and not doing real heavy lifting) If you're an enterprise programmer working on locked down networks and devices with more end point protection tools than endpoints, an unencumbered authorized system with access to the right network assets, as well as standardized and project specific tooling is the difference between coding and coding with one arm tied behind your back and a blindfold on. (actually the benefits for enterprise are many and not just DevEx -- think security, network admin, vendor access, governance, AI rollout, reduction in project specific on-boarding documentation....) I'm sure there are many other use cases, though these are just the immediate ones that spring to mind.
  - BOOSTERHIDROGEN 2y ago
    I appreciate the commentary; it provided a helpful overview of the use case.
TheTaytay 2y ago
Lunni looks cool. I hear mixed things about Docker Swarm. How has it been to work with?
- notpushkin 2y ago
  Thank you so much! Personally, Docker Swarm works great for me. I was worried about it for a while, too, but Mirantis seems to maintain it pretty well. There are even new features being shipped, like CSI support back in 2023! I’m thinking about adding a Kubernetes backend though, just in case things get worse for Swarm.
indigodaddy 2y ago
Lunni looks neat. Is persistence baked in with the docker stuff you deploy on and/or how is persistence handled/achieved?
- notpushkin 2y ago
  Thank you! We use the standard Docker volumes [1] for persisting data. It’s pretty straightforward – you just define a volume and specify where to mount it: volumes: app-data: {} services: app: image: example.com/app/backend:latest volumes: - app-data:/data This should work pretty well for most use cases. (There are some caveats if you decide to run multiple Lunni servers in a cluster – basically, every node would get its own volume by default – which is the main reason we don’t officially support that yet :-) [1]: https://docs.docker.com/engine/storage/volumes/ https://docs.docker.com/engine/storage/volumes/