Backend

/backend9190

Discussions about backend development

834

Gabriel Ayuso

@gabrielayuso.eth·

SQLite building a remote sync tool is a great move.

https://sqlite.org/draft/rsync.html

I don't know why I need to add members to /backend - I already had very little time to manage the channel. So, I added a few mods (that were mods before they got rid of mods and now mods are back) and if you want to be added to the channel, just reply here or DC one of us.

There's finally a growing discussion about moving away from PaaS into simple deployable fullstack services (thanks to @dhh of course).

I'm so here for it. We have been deploying our services to bare metal using https://kamal-deploy.org/ since day 1

If you want to send messages like this, /courtyard may be the place for you

I gotta say, as boring as it sounds, I thoroughly enjoy creating GraphQL schema and mapping it to my EdgeDB functions.

I always like the content Matt puts out about go. https://www.bytesizego.com/blog/golang-time

Do you prefer ArgoCD or fluxCD and why?

https://github.com/argoproj/argo-cd

https://github.com/fluxcd/flux2

This channel has rotted :(

This is a great listen for anyone who works on backend.

https://pca.st/episode/e51d97d3-7a11-4814-bbb3-1e29c825784f

One thing I never liked about many Apache projects is that they're written in JVM languages. I don't get doing system development in Java.

I guess it made more sense years ago and JVM has been turned significantly but still.

I hope no one seriously does system development on NodeJS. Bun might be a more performant option when it matures but I'd still stay away from any JS/TS runtime for system stuff.

Oh man I forgot about this channel. I've been doing /dev exclusive for a while now. I blame streaks.

Even though I ended my streak on /dev a while back. I got used to just casting in that channel.

Do you use buildpacks?

Any experience with buildpacks vs Docker?

https://buildpacks.io/

Have folks here played with NATS? https://www.youtube.com/watch?v=hjXIUPZ7ArM

This week we found out during some load tests that our metrics library (cloudwatch) could not keep up with the requests, eventually OOMing the pods. This is an inherent issue with push-based metrics vs pull-based ones (e.g. Prom). With pull-based metrics you do zero outbound API calls, just have the scraper hit your metrics endpoint at a rate that you control. Prometheus ftw!

Let's say I'll be tinkering with dotnet https://www.techempower.com/benchmarks/#section=data-r22&hw=ph&test=fortune

Incredible home cluster running ecommerce production. Would recommend to read some of the comments.

https://www.reddit.com/r/selfhosted/comments/1dqq3h8/comment/lapril8/

I was today years old when I learned

Every web framework needs this https://laravel-news.com/laravel-debugbar

This is great. My preferred stack is pretty similar, swapping django for phoenix. https://anthonynsimon.com/blog/one-man-saas-architecture/

After a year in production, I can confidently say: GraphQL subscriptions suck. We have much better tools to do the same thing, cleaner, faster, and more efficiently.

Does anyone have recommendations for a good SIMPLE logging service? I just want to be able to put some version of a console.log and via it in some hosted dashboard somewhere else.

I literally do not care about uptime monitoring or anything else.

A hill I'll die on:

nobody would write init_o_auth and a_p_i_endpoint, so miscapitalizing initialisms and midcap words in your javascript variables is wrong (ideal: initOauth, apiEndpoint)

If you're not following Marc's blog, you should https://www.brooker.co.za/blog/2024/06/04/scale.html

Tryna be productive, changing IPs on my domains, wondering how long my domain will resolve to the server, get hit with Linode's "we're undergoing maintenance" message.

SIGH

wow, 7 months later and I'm _again_ struggling with Cloudflare enablement hell. I love the product, but getting the team on the phone to pay them more money is shockingly hard

/Backend