API access via python script

DanScho · ‎12-18-2024

Hello,

I try to access an API via a python script.

I have access using postman. But if I try to access it via python in databricks, it will not work (timeout).

Here is my code:

import requests

import socket

url = 'https://prod-api.xsp.cloud.corpintra.net/tips-ki/symptom-prediction/api/v0/predict'

headers = {

'Content-Type': 'application/json',

'x-api-key': '<api-key>',

'User-Agent': 'PostmanRuntime/7.43.0',

'Accept': '*/*',

'Host': '<URL to host>',

'Accept-Encoding': 'gzip, deflate, br',

'Connection': 'keep-alive',

'Content-Length': '105'

}

payload = {

"result_l": "en",

"result_n": 2,

"texts": [

"Text1.",

"Text2"

]

}

response = requests.post(url, headers=headers, json=payload, timeout=10) # Timeout nach 10 Sekunden

response.raise_for_status() # Löst eine Ausnahme bei HTTP-Fehlern aus

data = response.json()

print(data)

Best regards,

Daniel

Walter_C · ‎12-18-2024

Are you running this from a notebook? if yes can you run bash command:

%sh

nc -vz prod-api.xsp.cloud.corpintra.net 443

I want to confirm your cluster is able to reach the endpoint.

DanScho · ‎12-18-2024

yes, i use it from a notebook.

C:\Windows>nc -vz prod-api.xsp.cloud.corpintra.net 443
Der Befehl "nc" ist entweder falsch geschrieben oder
konnte nicht gefunden werden.

DanScho · ‎12-18-2024

is "nc" a linux command? I use windows...

Walter_C · ‎12-18-2024

Sorry I am confused, you mentioned that you are running this in a notebook, this inside Databricks and running it through a cluster? if this is the case the command needs to be ran inside the notebook, you can add a new cell from it.

DanScho · ‎12-18-2024

sorry for confusing... i missunderstood your question.... now I got it.

running this command in a notebook in databricks shows no result.... it is "loading" endlessly... I only can interrupt it....

DanScho · ‎12-18-2024

now it shows:
nc: connect to prod-api.xsp.cloud.corpintra.net (53.13.186.164) port 443 (tcp) failed: Connection timed out

Walter_C · ‎12-18-2024

ok that means that your subnet for clusters in the cloud deployment is not allowing connections to this host on port 443, you will need to check with your cloud administrator or network team to allow this traffic

DanScho · ‎12-18-2024

hmm... not good... is it not possible to configure this in azure databricks by myself?

Walter_C · ‎12-18-2024

Unfortunately no because this is not being blocked by Databricks, this is being blocked by a firewall or security group at the cloud level

View solution in original post

DanScho · ‎12-19-2024

Nevertheless: Thank you very much for helping me!