search

swanchain / go-computing-provider

A golang implementation of computing provider
MIT License
11 stars 15 forks source link

rebooting #11

Closed ThomasBlock closed 5 months ago

ThomasBlock commented 5 months ago

I just wanted to ask what is the expected behaviour if i have to do maintenance

  1. how can i see which tasks / pods are running on which physical machine? and see how much cores etc are free?
  2. when i reboot a machine, all the tasks are just canceled and deleted. is this expected behaviour?
  3. how can i avoid Point 2? I can list all tasks with computing-provider task list -v and then kill every task with computing-provider task delete. is this oaky from a network perspective? will i get rewards ( partial until now) and my deposit back?
Normalnoise commented 5 months ago

I just wanted to ask what is the expected behaviour if i have to do maintenance

  1. how can i see which tasks / pods are running on which physical machine? and see how much cores etc are free?
  2. when i reboot a machine, all the tasks are just canceled and deleted. is this expected behaviour?
  3. how can i avoid Point 2? I can list all tasks with computing-provider task list -v and then kill every task with computing-provider task delete. is this oaky from a network perspective? will i get rewards ( partial until now) and my deposit back?

1\ you can run curl -k --location 'https://<YOUR_PUBLIC_IP>:<PORT>/api/v1/computing/cp'

2\ yes, Orchestrator will monster your service, if the service is not available, the task will be terminated.

3\ You must keep your server running, if the task is deleted, CP just can get a part of the total reward