Service status

This is where you land after signing in to a server, and it answers one question: is the service on this machine healthy?

What the page shows

• Running, stopped, failed to start, or not deployed — and when it stopped on its own, why, with the log from around that moment.
• CPU, memory and uptime, refreshed every second.
• Disk — sync storage, database, other data and free space, added up to the disk’s size.
• Storage, database size, vaults, users and devices, read every 30 seconds.
• Restarts — how often the machine has restarted the service. A number that keeps growing means it is crashing repeatedly.

When clients cannot connect

The page says which of these it is:

• The service port is not open. Open TCP 8443 (or the port you chose) in the cloud server’s firewall; port 80 has to be open too, because it renews the HTTPS certificate.
• The port answers, HTTPS is not ready. Usually the certificate is still being issued; wait a minute and click Check again.
• The admin interface is not responding. The numbers below it come from the last successful read.

The service logs

Click Service logs. It shows the most recent 1000 lines, filtered to All, Warnings above or Errors above. Export writes every log file the service keeps; Copy copies what is on screen.

Start, stop and delete

• Stop service pauses syncing on every device. Nothing is deleted, and More → Start service brings it back.
• Delete service stops the service and deletes the program, database, logs and backups under ~/courier; the certificate and its keys stay, so deploying to the same address again reuses them. Every connected plugin stops syncing at once, and the data has no undo, so use it when the machine is being retired, not to pause.

The More menu also holds Redeploy (to change the port or the program) and Moving to a new server.