`prefect_aws.microvm.server`

HTTP server for the AWS Lambda MicroVM lifecycle hooks.

Classes

`LifecycleHookServer`

Serves the MicroVM lifecycle hooks and hands the first valid /run request to the thread that calls wait_for_run_request. Hooks are accepted both at /<hook> and at /aws/lambda-microvms/runtime/v1/<hook>. Methods:

`claim_run`

claim_run(self) -> bool

Return True for the first /run request and False for every later one.

`process_request`

process_request(self, request: Any, client_address: Any) -> None

`process_request_thread`

process_request_thread(self, request: Any, client_address: Any) -> None

`server_bind`

server_bind(self) -> None

`stop_serving`

stop_serving(self, timeout: float = HANDLER_FINISH_TIMEOUT_SECONDS) -> None

Stop accepting hook requests and wait up to timeout seconds for the requests in progress to finish, so that no handler thread holds a lock when the process forks. Requests still in progress after timeout are logged and left running. Args:

timeout: The longest time to wait for requests in progress, in seconds.

`submit_run`

submit_run(self, request: RunHookRequest | InvalidRunHookPayload) -> None

Pass a parsed /run request, or the reason it was rejected, to wait_for_run_request.

`wait_for_run_request`

wait_for_run_request(self) -> RunHookRequest | InvalidRunHookPayload

Block until the first /run request has been answered, then return it.

`LifecycleHookHandler`

Request handler for LifecycleHookServer. Methods:

`do_GET`

do_GET(self) -> None

`do_POST`

do_POST(self) -> None

`log_message`

log_message(self, format: str, *args: Any) -> None