TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Ask HN: Hosting on Digital Ocean, any advice for monitoring and deployments?

19 点作者 hershyb_6 个月前
I'm moving over from Lambda to Digital Ocean since I found it much easier to test locally and iterate. Unfortunately, I'm not too savvy with setting up monitoring so I can see how often failures are happening and how to smoothly deploy changes on DO. I currently have a setup where I pull each individual change into my instance and re-run docker-compose build etc. Would love to learn more about what tools have helped you for logging/monitoring/deployments

9 条评论

Hammad_khaan6 个月前
Think of your app like a video game—tools like Prometheus and Grafana are your scoreboard and stats screen, showing what's up with your game. For smoother moves (deployments), try GitHub Actions or Jenkins; they’re like power-ups for easy leveling!
babyent6 个月前
I use DO as well and I use GitHub actions.<p>My backend is node-ts and a couple Go programs.<p>Via actions I can restart pm2, build Go, restart those services, and it works great.<p>As for monitoring, I’m just relying on the DO resource alerts. They’re fine for my needs.<p>Sorry if it’s too basic. Your needs may very well be different.<p>I’m trying to keep my setup as simple as possible since I’m solo right now. I used to be a platform engineer before and I don’t think I need anything fancier until I have a few hundred plus customers (b2b) and a dozen employees. Could be wrong, but it’s a full time job doing anything more imo.
评论 #42104831 未加载
_jsdp6 个月前
If your app is instrumented with OpenTelemetry, I&#x27;d really recommend something like SigNoz (<a href="https:&#x2F;&#x2F;signoz.io&#x2F;">https:&#x2F;&#x2F;signoz.io&#x2F;</a>) or Axum (<a href="https:&#x2F;&#x2F;axiom.co&#x2F;" rel="nofollow">https:&#x2F;&#x2F;axiom.co&#x2F;</a>) (or even <a href="https:&#x2F;&#x2F;baselime.io" rel="nofollow">https:&#x2F;&#x2F;baselime.io</a> but their application is a mess to work with)<p>1. Axiom has a generous free plan (not sure if it&#x27;s self-hostable) 2. SigNoz has an easy to work with self-hosting setup but also cloud solution.<p>If it&#x27;s just logging, you can still forward your logs to these platforms but having OTel really goes a long way (from debugging and vendor selection POV).
dengolius6 个月前
Take a look on <a href="https:&#x2F;&#x2F;docs.victoriametrics.com&#x2F;#prominent-features" rel="nofollow">https:&#x2F;&#x2F;docs.victoriametrics.com&#x2F;#prominent-features</a> and <a href="https:&#x2F;&#x2F;docs.victoriametrics.com&#x2F;victorialogs&#x2F;quickstart" rel="nofollow">https:&#x2F;&#x2F;docs.victoriametrics.com&#x2F;victorialogs&#x2F;quickstart</a>
lazypenguin6 个月前
Grafana cloud, install agents, pretty dashboard. Can add custom prom ingestor to load application logs. Sleep good. Feel confident, look at grafana alerts. Cry much. Finally learn promql, pro dashboards, sleep good again.<p>Seriously though grafana cloud with their agents is pretty easy and affordable. Grafana family of products kind of have a learning curve but they work very well.
评论 #42073696 未加载
raihansaputra6 个月前
On deployment side, Portainer seems to fit your needs. Can run arbitrary containers and docker-compose stacks, but unsure whether it can do image builds on the server.
satvikpendem6 个月前
For deployment, you could look at Coolify or Dokploy.
mochja6 个月前
<a href="https:&#x2F;&#x2F;www.netdata.cloud&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.netdata.cloud&#x2F;</a>
magundu6 个月前
Try <a href="https:&#x2F;&#x2F;atatus.com" rel="nofollow">https:&#x2F;&#x2F;atatus.com</a>