search

tensorchord / openmodelz

One-click machine learning deployment (LLM, text-to-image and so on) at scale on any cluster (GCP, AWS, Lambda labs, your home lab, or even a single machine).
https://docs.open.modelz.ai
Apache License 2.0
239 stars 23 forks source link

bug: 404 inference not found #175

Closed singeleaf closed 1 year ago

singeleaf commented 1 year ago

Here's the list info detail

NAME       ENDPOINT                                             STATUS  INVOCATIONS  REPLICAS 
nginx-ing  http://nginx-ing-yjmz0eabd9c97h08.modelz.live        Ready             0  1/1      
           http://localhost:80/inference/nginx-ing.test

Then when I access the deployment by using endpoint, the error message has been raised

curl  http://nginx-ing-yjmz0eabd9c97h08.modelz.live
{"http_status_code":404,"message":"inference not found","request":"GET /inference/nginx-ing.default/","op":"inference-proxy","error":{}}
VoVAllen commented 1 year ago

Hi,

It's weird and seems the ip address is missing. The link should looks like http://nginx-ing-yjmz0eabd9c97h08.<ip-address>.modelz.live

Could you try initialize mdz with server public ip?

# Provide the public IP as an argument
$ mdz server start 1.2.3.4