Choose a Postmark server, sender, and message stream
Set up Postmark's server token and authenticated sender, then choose a transactional stream before connecting the application.
Reviewed
Postmark organizes sending and inbound processing into servers. A server API token is not the same as the account token used to administer account-level resources. Choose the server and traffic stream deliberately before giving an application credentials.
Prepare the sending context
- Create or select the server appropriate to this application/environment.
- Verify the sender/domain and publish the authentication records shown in Postmark.
- Select the transactional message stream for application-triggered mail; keep Broadcast traffic separate.
- Store the server token in a backend secret manager. Do not grant account administration merely to send notifications.
Review current Postmark documentation for account approval and domain requirements. Sender verification does not configure inbound delivery automatically.
Verify the intended server
Run a controlled send with the expected stream and inspect the correct server’s activity. Check the envelope/from identity and stream in the log, not just the message’s visible subject.
If the request is rejected, distinguish a wrong token/server from an unverified sender or stream mismatch. Replacing the key with an account-wide credential is not a least-privilege fix.
Maintain separate development and production configurations. Postmark’s sandbox facilities can help inspect behavior, but a sandbox acceptance does not prove real recipient delivery. MailKite credentials and domain checks remain separate if you later change the delivery transport.