Ask HN: How do you trigger deployments in your workflow?
I’m trying to understand how teams (or solo developers) actually kick off deployments.
Not the runtime (K8s, VMs, ECS etc) ... but the hook/trigger/pipeline that rolls out code.
For example:
- Do you have a custom roll-your-own pipeline or IDP?
- Do you deploy from Git branches (e.g. push to `main` triggers a deploy)?
- Do you have scripts that run on a CI server?
- Webhooks that trigger a rollout on your infra?
- ChatOps commands?
- Manual steps like SSH’ing into a box and running `git pull`?
- Cron-style periodic syncs?
- Something else entirely?
I'm especially curious why you ended up with that trigger mechanism ... constraints, team size, simplicity, reliability needs etc.
How does your deployment trigger flow look these days?
I like to deploy automatically from pushes to dev, staging and main which are all gated by mandatory PR reviews, and any scripts run from Lambda which are gated by account access, so it's always a deliberate occurrence and only certain people can trigger it and nobody needs to be doing anything manually against the environments.