Serve CLI not working into ray==2.3.0

Ray Libraries (Data, Train, Tune, Serve) Ray Serve
1

serve status not working into ray>2.3.0.

2

@Ajay_Kumar Can you provide an example CLI and what’s the output you getting? For what you have is hard to go by?

cc: @Sihan_Wang @cindy_zhang

cheers, HTH
Jules

3

sever status showing app_status NOT_STARTED but the app is running. dashboard and API also working.

(ray_api) ubuntu@ubuntu-pc:~/projects$ serve status
name: ''
app_status:
  status: NOT_STARTED
  message: Application  doesn't exist
  deployment_timestamp: 0.0
deployment_statuses: []

Version
ray==2.3.0
python==3.10.0

deployment

(ray_api) ubuntu@ubuntu-pc:~/projects$ ray stop
Stopped all 7 Ray processes.
(ray_api) ubuntu@ubuntu-pc:~/projects$ ray start --head --num-cpus=64 --num-gpus=1 --dashboard-host 0.0.0.0
Usage stats collection is enabled. To disable this, add `--disable-usage-stats` to the command that starts the cluster, or run the following command: `ray disable-usage-stats` before starting the cluster. See https://docs.ray.io/en/master/cluster/usage-stats.html for more details.

Local node IP: 192.168.12.191

--------------------
Ray runtime started.
--------------------

Next steps
  To connect to this Ray runtime from another node, run
    ray start --address='192.168.12.191:6379'
  
  Alternatively, use the following Python code:
    import ray
    ray.init(address='auto')
  
  To see the status of the cluster, use
    ray status
  To monitor and debug Ray, view the dashboard at 
    192.168.12.191:8265
  
  If the connection fails, check your firewall settings and network configuration.
  
  To terminate the Ray runtime, run
    ray stop

(ray_api) ubuntu@ubuntu-pc:~/projects$ python main_copy.py
2023-05-02 11:34:47,308 INFO worker.py:1364 -- Connecting to existing Ray cluster at address: 192.168.12.191:6379...
2023-05-02 11:34:47,314 INFO worker.py:1544 -- Connected to Ray cluster. View the dashboard at 192.168.12.191:8265 
(ServeController pid=3529918) INFO 2023-05-02 11:34:47,939 controller 3529918 http_state.py:129 - Starting HTTP proxy with name 'SERVE_CONTROLLER_ACTOR:SERVE_PROXY_ACTOR-08d5632b43b5d4d01cf691d4c942fc48138fe49da3d069c72578f14b' on node '08d5632b43b5d4d01cf691d4c942fc48138fe49da3d069c72578f14b' listening on '0.0.0.0:8000'
(ServeController pid=3529918) INFO 2023-05-02 11:34:48,581 controller 3529918 deployment_state.py:1333 - Adding 1 replica to deployment 'Hello'.
(HTTPProxyActor pid=3529977) INFO:     Started server process [3529977]
(ray_api) ubuntu@ubuntu-pc:~/projects$ serve status
name: ''
app_status:
  status: NOT_STARTED
  message: Application  doesn't exist
  deployment_timestamp: 0.0
deployment_statuses: []

main_copy.py

from ray import serve
import starlette.requests

@serve.deployment(
    route_prefix= "/hello",
)
class Hello:
    def __call__(self, request: starlette.requests.Request):
        return "Hello, Greeting from API."

serve.start(
    detached=True,
    http_options={
        "host": "0.0.0.0",
        "port": "8000"
    }
)

Hello.deploy()
4

@Ajay_Kumar Seems like you using deprecated APIs. Use bind and serve.run(). Try and let us know. serve.run() has all the kwargs you need

cc: @cindy_zhang @Sihan_Wang