Skip to main content
Tries to recover a computer that is not answering, then checks it again. By default it never restarts the computer. If the computer is still not healthy, recommendation says what to do next. Send {"action": "restart"} to restart it. To look without changing anything, call Check computer health.

Path parameters

string
required
Computer ID (UUID). An instance_id is not accepted here.

Body parameters

string
default:"auto"
auto tries automatic recovery and respects a computer you paused. restart restarts the computer, then waits up to about 45 seconds for it to answer. Any other value reads as auto. The body is optional.

Example

Response

string
The computer UUID.
string
auto or restart.
boolean
true when the computer is healthy after the call.
object[]
What was done, in order, each with step and detail.
object
The checks after the call, as Check computer health returns them.
string
What to do next. Present when action is auto and healed is false, for example to resume a paused computer or to call this endpoint again with {"action": "restart"}.
string
Present when a wake from storage did not finish. computer_waking means it is still waking, so check again in a minute. Any other value is the reason it failed.
number
Seconds to wait before you ask again, when there is a time to give.

Errors