Hi Prefect experts am trying to deploy a workflow on prefect

Question

Hi Prefect experts am trying to deploy a workflow on prefect cloud setup on our Kubernetes cluster when I run `prefect auth create token n my runner token s RUNNER` i get this error

Ajith Kumara Beragala Acharige Lal · Answer

`Traceback most recent call last ` `File home ajith local lib python3 8 site packages prefect client client py line 451 in request` `json resp = response json ` `File usr lib python3 dist packages requests models py line 897 in json` `return complexjson loads self text kwargs ` `File usr lib python3 dist packages simplejson init py line 518 in loads` `return default decoder decode s ` `File usr lib python3 dist packages simplejson decoder py line 370 in decode` `obj end = self raw decode s ` `File usr lib python3 dist packages simplejson decoder py line 400 in raw decode` `return self scan once s idx= w s idx end ` `simplejson errors JSONDecodeError Expecting value line 1 column 1 char 0 ` `The above exception was the direct cause of the following exception ` `Traceback most recent call last ` `File home ajith local bin prefect line 8 in lt module gt ` `sys exit cli ` `File usr lib python3 dist packages click core py line 764 in call ` `return self main args kwargs ` `File usr lib python3 dist packages click core py line 717 in main` `rv = self invoke ctx ` `File usr lib python3 dist packages click core py line 1137 in invoke` `return process result sub ctx command invoke sub ctx ` `File usr lib python3 dist packages click core py line 1137 in invoke` `return process result sub ctx command invoke sub ctx ` `File usr lib python3 dist packages click core py line 956 in invoke` `return ctx invoke self callback ctx params ` `File usr lib python3 dist packages click core py line 555 in invoke` `return callback args kwargs ` `File home ajith local lib python3 8 site packages prefect cli auth py line 214 in create token` `output = client graphql ` `File home ajith local lib python3 8 site packages prefect client client py line 298 in graphql` `result = <http self post|self post> ` `File home ajith local lib python3 8 site packages prefect client client py line 213 in post` `response = self request ` `File home ajith local lib python3 8 site packages prefect client client py line 454 in request` `raise ClientError ` `prefect utilities exceptions ClientError Malformed response received from Cloud please ensure that you have an API token properly configured `

Ajith Kumara Beragala Acharige Lal · Answer

setup is like this

Ajith Kumara Beragala Acharige Lal · Answer

agents registered

Ajith Kumara Beragala Acharige Lal · Answer

any idea why i cannot create a token on this cloud

Chris White · Answer

Hi Ajith this looks like you are mixing backends Prefect Server does not have any concept of users or tokens

Ajith Kumara Beragala Acharige Lal · Answer

Hi < Chris White> thanks if I understand your answer correct We cannot setup cloud mode on docker only the server mode

Ajith Kumara Beragala Acharige Lal · Answer

< Chris White> and I tried to run a simple ShellTask after switching back to `server` then i get this error

Ajith Kumara Beragala Acharige Lal · Answer

```requests exceptions ConnectionError HTTPConnectionPool host= localhost port=4200 Max retries exceeded with url graphql Caused by NewConnectionError lt urllib3 connection HTTPConnection object at 0x7f4fcbcc5130 gt Failed to establish a new connection Errno 111 Connection refused ```

Chris White · Answer

No that s not what I mean Prefect consists of two components Prefect Core the workflow engine builder that you write your workflows with and test them run them locally A Prefect Backend API this is an API that your workflows can be configured to talk to this API is where persistence of state and other interesting orchestration features happens You have two mutually exclusive options for the API you can self host the open source version of the API Prefect Server and in doing so you need to make sure both your agents and the workflows they submit have network access to this API that you have stood up Alternatively you can use Prefect Cloud which is the API that we manage Your issues are resulting from either trying to use Cloud settings with a Server API or vice versa or because your agents flows do not have network access to the Server API that you have created

Ajith Kumara Beragala Acharige Lal · Answer

Thank you very much < Chris White> sorry to bugging you if agent is on docker how can I register a flow i have prior experience when agent is running on a machine instead docker any example or documentation

Chris White · Answer

No worries at all I totally understand your confusion there are a lot of moving pieces and we re definitely trying to streamline our documentation to make these things easier The key thing is that you need to make sure the docker containers have network access to the API give me a minute and I can try to find some docs an example for you

Ajith Kumara Beragala Acharige Lal · Answer

Thanks a lot < Chris White> if I m not mistaken agent and server has established the communication

Ajith Kumara Beragala Acharige Lal · Answer

I guess i don t know where to execute and register the script

Ajith Kumara Beragala Acharige Lal · Answer

hi < Chris White> It d be really great if you can give some reference today as the my Jira task related to this for this almost met the deadline pray pray pray pray

Chris White · Answer

Hi < Ajith Kumara Beragala Acharige Lal> it looks like you are running the kubernetes agent which should be much simpler because Kubernetes should be handling the networking heavy lifting here is a helm chart template for the k8s agent against a server deployment that you might want to follow For your flow I recommend using `Docker` storage in this particular instance with a registry that your k8s cluster has access to

Ajith Kumara Beragala Acharige Lal · Answer

Thanks a lot < Chris White> i ll get back to you with results Good night

Chris White · Answer

For sure good luck