Welcome • User Guide
ToS • Privacy • Canary
Donate • Bugs • License

©2025 Poal.co

631

Archive: https://archive.today/jKGQX

From the post:

>Or: How I learned that “I don’t use Next.js” doesn’t mean your dependencies don’t use Next.js

Archive: https://archive.today/jKGQX From the post: >>Or: How I learned that “I don’t use Next.js” doesn’t mean your dependencies don’t use Next.js

(post is archived)

[–] • 1 pt

No way, it was running on docker....

I said this a while back....and again, and again.... Docker is just someone elses unknown linux build. Get fucked cunt.

[–] • 2 pts

FFS we talked about that this week.

Fucking moron running a fucking docker built by a fucking idiot that has no fucking clue what they are doing exposed to the fucking internet.

Not only that, the dumbass didn't have monitoring on his system(s). If my system spikes to 800% fucking CPU I am going to be notified. I would have caught that in minutes or less and shut it the fuck down while I take apart the (offline) image to see what the fuck happened.

STOP LETTING DEV'S DEPLOY ANYTHING.

I AM SO TIRED OF THIS STUPID.

[–] • 1 pt

...as I sit here and hand bulid a linux box, and its totally inside my shitwall.....but still...do shit the old way.

I wish for Ubuntu Jeos back.....I loved that shit. But now we have phased updates and pay as you go security patches.

Devaun enters the chat.

yeah devaun, I'll get back to you in a few

[–] • 0 pt

I automate my build process to create a master image that is easily reproducible.

Yes, even for what I deploy at home. I want everything to have a base set of tools, typical mount points, etc....

Then I create roles/automation around what specific task that system is designed for so it is also reproducible. What if I need to rebuild it from scratch? well, its all in my automation and if there are bugs I can figure them out quickly at run-time and a system rebuild that might take a couple days now is a hour or two at most (usually less than a hour since the automation works).

This is all in a local repo backed up to a remote site fully encrypted so I can retrieve it if my hardware fails... etc... I don't quite have that redundancy for (all) of my data but I am working on it.

Eventually if I need to I could fully rebuild everything I run at home by running a couple of jobs after spending ~10 minutes to stand up a master-node to kick the process off.

I have "built datacenters" with a config file and a terminal/node/whatever you want to call it. Its not that hard when you take apart all of the pieces and make them small.