Sprite non-responsive

Questions / Help
1

I have a long running agents working on my codebase and I fear all those tokens are used for nothing

Error: failed to list sessions: failed to list sessions: Get “https://api.sprites.dev/v1/sprites/remoto-2/exec”: net/http: request canceled (Client.Timeout exceeded while awaiting headers)

Error: failed to list checkpoints: failed to make request: Get “https://api.sprites.dev/v1/sprites/remoto-2/checkpoints”: net/http: request canceled (Client.Timeout exceeded while awaiting headers)

Error: failed to start sprite command: failed to connect: read tcp xxx->xxx: i/o timeout

Is there anyway to restart a hung sprite?

3

Hey!

I want to say you should be able to restart the unresponsive Sprite using the machine level restart endpoint, if I’m not mistaken:

curl -X POST \
  -H "Authorization: Bearer $SPRITE_TOKEN" \
  "https://api.sprites.dev/v1/sprites/remoto-2/restart"

This is different from /services/{service_name}/restart, which only restarts one managed service inside a reachable Sprite. Since exec, session, and checkpoint requests are all timing out, the full Sprite restart is the appropriate recovery action here. The restart should reboot the Sprite and terminate the agent process that is currently running. Files and repo changes already written to the Sprite’s persistent filesystem should remain after it comes back, but unsaved or in-memory work might be lost.