[KongchangAI]
ProductVercel Daemons

Daemons

Long-running background processes on the Vercel platform whose lifecycle is bound to a deployment. They are suited to background logic such as workers, queue consumers, and scheduled tasks, and can access other services within the same deployment through service bindings. Unlike request-driven units, daemons persist beyond individual invocations, making them appropriate for continuous processing and asynchronous workloads that must remain active for the duration of a deployment.

Timeline (last 90 days)

Oct 30

Vercel introduced service bindings for Daemons, allowing resident background daemons to directly call other services within the same deployment.

Unverified50%
Oct 30

In Vercel's deployment model, Services are request-driven application units that typically respond to calls as HTTP services; Daemons are resident processes whose lifecycle is bound to the deployment, suited to background logic such as workers, queue consumers and scheduled tasks.

Unverified50%
Oct 30

Before service bindings, communication between a Daemon and a Service relied on externally exposed HTTP addresses, requiring either making the service public or building a custom internal access channel.

Unverified50%
Oct 30

With service bindings, a Daemon can declare a binding to obtain the target service's URL, and Vercel handles internal routing, authentication and TLS, while the service itself remains private and does not need to be exposed externally.

Unverified50%
Oct 30

Once the Daemon has the injected URL, it can make requests using standard fetch, without importing a proprietary SDK or handling signatures/certificates.

Unverified50%
Oct 30

Service bindings are one-way: Services can bind to services outside Daemons, but Services cannot bind to Daemons.

Unverified50%
Oct 30

Each Daemon instance always calls the services bound to the deployment that started it, even if a promotion or rollback occurs afterward.

Unverified50%

All Facts (7)

Source Articles