Skip to content
Changelog

Deployments

Every deploy, checked

Vercel, Netlify, Cloudflare or your own server tell PostShip, which checks every page within the second, then at T+2 and T+8.

Most outages happen right after a release. So PostShip checks when it counts: as soon as your host announces a deploy, not at the probe's next run.

A commit goes to the host; the webhook tells PostShip within the second, which checks every page and gives a verdict.

Told within the second

Vercel, Netlify and Cloudflare Pages connect directly, previews included. For everything else — Coolify, Railway, Render, a VPS, a CI job — a generic webhook, whose secret travels in the X-PostShip-Secret header and never in the address. The same release received twice doesn't rerun the checks.

Checked, then checked again

Every production deploy is checked straight away, then again at T+2 and T+8 minutes, for what only breaks once the cache and CDN have settled. PostShip also compares what your domain serves with the deploy's address: if the domain still loads the old version twenty seconds later — cache, alias, DNS — you are warned. And when your host's firewall sets an anti-bot challenge right after the release, the check concludes nothing instead of reporting an outage.

The history

The Deployments page keeps every release with its commit — message, branch, author —, the build log at Vercel, errors in red, and what broke or recovered since the previous one. Each deploy's share card is archived as it goes out: weeks later, you can still show what “before” looked like.

Read the documentation