Wrong API Key Spawns Phantom Inference Pipeline Process #907
Comments
|
Hi, is anyone working on this? I'd love to make a contribution if this is still unassigned |
|
Hi @farhan0167 , I have not started working on this issue yet so please feel free to give it a try! Thanks! |
|
hi @yeldarby , so I tried running the minimum reproducible as is: from inference_sdk import InferenceHTTPClient
client = InferenceHTTPClient(
api_url="http://localhost:9001", # use local inference server
api_key="invalid" # optional to access your private data and models
)
result = client.start_inference_pipeline_with_workflow(
video_reference=[0],
workspace_name="roboflow-docs",
workflow_id="clip-frames",
max_fps=max_fps,
workflows_parameters={
"prompt": "blurry",
"threshold": 0.16
}
)and I'm getting the following error which I believe is what you are expecting: Environment
Is there anything that I'm missing that I could try out? Is there another public workflow that I could try out that gives you the error? |
|
Hi @farhan0167 , do you get more output? @yeldarby suggests we should see infinite chain of lines like the one below: |
|
@grzegorz-roboflow yeah I actually tried some different examples, but I wasn't able to reproduce the error with an infinite chain of logs. If i don't provide an api key for the following for example: Then i get a normal response back but when i uncomment the line, and re-run it, I get only: |
|
Hi @farhan0167 , I fixed the infinite loop in #983 (please have a look if you'd like to understand how I reproduced the error), I think we still have good area for improvement here when it comes to returned error, would you like to look into this? |
|
I'd be interested in taking a look at this. And I see why I wasnt seeing the infinite loop. I was actually not looking at the logs for the server. Before I get started, could you explain how you got your webcam running? Having a hard time running my webcam using the example here from inference_sdk import InferenceHTTPClient
import atexit
import time
max_fps = 4
client = InferenceHTTPClient(
api_url="http://localhost:9001", # use local inference server
#api_key="" # optional to access your private data and models
)
# Start a stream on an rtsp stream
result = client.start_inference_pipeline_with_workflow(
video_reference=0,
workspace_name="test-coqbp",
workflow_id="detect-and-classify",
max_fps=max_fps,
workflows_parameters={
"prompt": "blurry", # change to look for something else
"threshold": 0.16
}
)I get something like this: before it stops. Not sure if there's an issue that already addresses this or if this should be its own issue. But as soon as I understand how everything's working, I can look into the error thing. |
|
Hi @farhan0167, I start inference pipeline container: docker build \
-t roboflow/roboflow-inference-server-cpu:test \
-f docker/dockerfiles/Dockerfile.onnx.cpu \
.
docker run \
--rm \
-p 9001:9001 \
-p 9002:9002 \
-e NOTEBOOK_ENABLED=true \
-e ENABLE_STREAM_API=true \
-e LOG_LEVEL=DEBUG \
-e LMM_ENABLED=True \
--name=hosted_inference \
-v ./inference:/app/inference \
-v ./docker/config/cpu_http.py:/app/cpu_http.py \
roboflow/roboflow-inference-server-cpu:testThen, in my browser, I visit Hope this helps! |
|
The original issue of pipeline not terminated was resolved, I'm reopening this since we can add improvement related to this issue - when pipeline fails we can send more descriptive error message back to the UI |
Search before asking
Bug
If you pass the wrong API key to
/inference_pipelines/initialiseit will create a zombie process that spits out all sorts of errors indefinitely.Logs:
Environment
Minimal Reproducible Example
Using a public Workflow that doesn't require an API Key but passing an invalid one:
Additional
No response
Are you willing to submit a PR?
The text was updated successfully, but these errors were encountered: