Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Custom timeouts might not be used with REST api #874

Closed
Villux opened this issue Sep 18, 2019 · 1 comment
Closed

Custom timeouts might not be used with REST api #874

Villux opened this issue Sep 18, 2019 · 1 comment
Assignees
Milestone

Comments

@Villux
Copy link

Villux commented Sep 18, 2019

This is my seldon configuration:

apiVersion: machinelearning.seldon.io/v1alpha2
kind: SeldonDeployment
metadata:
    labels:
        app: seldon
    name: "model-{{workflow.parameters.model-id}}"
spec:
    annotations:
        deployment_version: v1
        project_name: Cat Example Pytorch
        seldon.io/rest-read-timeout: '10000000000'
        seldon.io/rest-connection-timeout: '10000000000'
        seldon.io/rest-connect-retries: "1"
        seldon.io/grpc-read-timeout: "10000000000"
...

With these super long timeouts I still get '{\n "code": 203,\n "info": "Failed to retrieve predictions after 3 attempts",\n "reason": "Microservice error",\n "status": "FAILURE"\n}' from my client.

Pod containing the rest api has this logging

i.s.engine.config.AnnotationsConfig      : Annotations {kubernetes.io/config.source=api, kubernetes.io/config.seen=2019-09-18T14:26:47.596964209Z, prometheus.io/path=prometheus, prometheus.io/port=8000, prometheus.io/scrape=true}
i.seldon.engine.tracing.TracingProvider  : Not activating tracing
i.s.e.service.InternalPredictionService  : REST Connection timeout set to 200
i.s.e.service.InternalPredictionService  : REST read timeout set to 5000
i.s.e.service.InternalPredictionService  : gRPC max message size set to 4194304
i.s.e.service.InternalPredictionService  : gRPC read timeout set to 5000
i.s.e.service.InternalPredictionService  : REST retries set to 3
io.seldon.engine.grpc.SeldonGrpcServer   : FOUND env var [ENGINE_SERVER_GRPC_PORT], will use engine server port 5001

spring outputs this error: org.springframework.web.client.ResourceAccessException: I/O error on POST request for "http://url.kubeflow.svc.cluster.local.:9000/predict": Read timed out; nested exception is java.net.SocketTimeoutException: Read timed out

If I read it correctly it's not using the timeout values that I have specified. Do you have any idea why it might be so? I'm pretty sure that I have a configuration mistake, but I can't find anything from the document that would differ from this.

@axsaucedo axsaucedo added this to the 1.1 milestone Jan 16, 2020
@ukclivecox ukclivecox self-assigned this Jan 31, 2020
@ukclivecox
Copy link
Contributor

Should be fixed in 1.1 with new executor.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants