← 返回首页
Add support for enabling metrics in Feast Operator · Issue #5317 · feast-dev/feast · GitHub
Skip to content

Navigation Menu

Toggle navigation
Sign in
Appearance settings
Search or jump to...

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Include my email address so I can be contacted

Saved searches

Use saved searches to filter your results more quickly

Appearance settings
Resetting focus

Add support for enabling metrics in Feast Operator #5317

New issue
New issue

Description

Is your feature request related to a problem? Please describe.
I'm trying to enable metrics in my Feast deployment using the Feast Operator, but there's no clear way to pass the --metrics flag to the online container. The documentation provides examples for Helm chart deployment but not for Operator-based deployments. When attempting to use a custom image with an entrypoint that includes the --metrics flag, I found that the Operator doesn't respect this entrypoint configuration.

Describe the solution you'd like
Add a configuration option in the Feast Operator specification to enable metrics for the online server. Ideally, this would be a simple boolean flag in the CustomResource definition that, when set to true, would add the --metrics flag to the command arguments of the online container.

Describe alternatives you've considered
I've tried creating a custom Docker image with an ENTRYPOINT that includes the --metrics flag, but the Operator seems to override this with its own command. Here's my Dockerfile:

FROM quay.io/feastdev/feature-server:0.49.0 # Install ddtrace with OpenTelemetry support RUN pip install 'ddtrace[opentelemetry]>=1.0.0' # Add environmental configurations ENV DD_TRACE_ENABLED=true ENV DD_LOGS_INJECTION=true ENV DD_RUNTIME_METRICS_ENABLED=true # Use feast command directly as the entrypoint with metrics flag ENTRYPOINT ["feast", "serve", "-h", "0.0.0.0", "-p", "6566", "--metrics"]

Additional context

When examining the pods created by the Operator, I can see that my custom entrypoint is being ignored. The Operator is setting its own command:

Containers: online: Container ID: containerd://f6c09abfd39d85a60bba9bc5290eca07a0c8b5120ab2e6fc16ee2676cd59d18f Image: us-central1-docker.pkg.dev/artifacts/vertex-services-docker/feast-staging:sha-41c10 Image ID: us-central1-docker.pkg.dev/artifacts/vertex-services-docker/feast-staging@sha256:0956b2af3fc7bc6cca143f3a886466989d8bb0827fb2a0f1e60d26990fe4d744 Port: 6566/TCP Host Port: 0/TCP Command: feast --log-level DEBUG serve -h 0.0.0.0 -p 6566

The Helm chart deployment supports metrics through --set metric=true, but there doesn't seem to be an equivalent option for the Operator. Clear documentation on how to enable metrics with the Feast Operator would also be helpful.

Metadata

Metadata

Assignees

No one assigned

    Labels

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions

      Footer

      © 2026 GitHub, Inc.