Skip to content

Commit

Permalink
Update the example Dockerfile to run on k8s (#5412)
Browse files Browse the repository at this point in the history
Signed-off-by: Jason Parraga <[email protected]>
  • Loading branch information
Sovietaced authored May 22, 2024
1 parent 2f7bedf commit 317ad30
Showing 1 changed file with 4 additions and 5 deletions.
9 changes: 4 additions & 5 deletions docs/flyte_agents/developing_agents.md
Original file line number Diff line number Diff line change
Expand Up @@ -140,14 +140,13 @@ You can test your agent in a {ref}`local Python environment <testing_agents_loca
The following is a sample Dockerfile for building an image for a Flyte agent:

```Dockerfile
FROM python:3.9-slim-buster
FROM python:3.10-slim-bookworm

MAINTAINER Flyte Team <[email protected]>
LABEL org.opencontainers.image.source=https://github.com/flyteorg/flytekit

WORKDIR /root
ENV PYTHONPATH /root

# additional dependencies for running in k8s
RUN pip install prometheus-client grpcio-health-checking
# flytekit will autoload the agent if package is installed.
RUN pip install flytekitplugins-bigquery
CMD pyflyte serve agent --port 8000
Expand Down Expand Up @@ -193,7 +192,7 @@ By running agents independently, you can thoroughly test and validate your agent
controlled environment before deploying them to the production cluster.

By default, all agent requests will be sent to the default agent service. However,
you can route particular task requests to designated agent services by adjusting the FlytePropeller configuration.
you can route particular task requests to designated agent services by adjusting the FlytePropeller configuration.

```yaml
plugins:
Expand Down

0 comments on commit 317ad30

Please sign in to comment.