enable setting the working directory when using image spec about flyte

[dosubot]

🤖

Hello @samhita-alla,

I'm Dosu, a friendly bot here to assist you with your questions, bugs, and help you on your journey to becoming a contributor while we wait for our human maintainers. I'm here to make your experience with the flyte repository as smooth as possible. Let's get started!

Sure, you can set the working directory when using image spec in Flyte. This can be done in your Dockerfile using the WORKDIR directive. This sets the working directory for any RUN, CMD, ENTRYPOINT, COPY, and ADD instructions that follow in the Dockerfile. If the WORKDIR doesn’t exist, it will be created even if it’s not used in any subsequent Dockerfile instruction.

Here's an example of how you can modify your Dockerfile:

FROM --platform=${BUILDPLATFORM} golang:1.21.5-alpine3.18 as builder

# ... rest of your Dockerfile ...

# This will eventually move to centurylink/ca-certs:latest for minimum possible image size
FROM alpine:3.18
LABEL org.opencontainers.image.source https://github.com/flyteorg/flytepropeller

COPY --from=builder /artifacts /bin

RUN apk --update add ca-certificates

RUN addgroup -S flyte && adduser -S flyte -G flyte
USER flyte

WORKDIR /root

CMD ["flytepropeller"]

In this example, WORKDIR /root sets the working directory to /root for the CMD ["flytepropeller"] instruction and any other instructions that follow.

Regarding how Flyte handles the working directory for different tasks and workflows, it provides a mechanism for tasks to access their inputs and outputs through specific paths. This is done through the DataLoadingConfig message in the Container and K8sPod messages. The input_path and output_path fields specify the file system paths where the inputs and outputs for the task are located. When a task is executed, the inputs are loaded into the input_path, and the task is expected to write its outputs to the output_path.

Here is the relevant part of the proto file:

message DataLoadingConfig {
    // Flag enables DataLoading Config. If this is not set, data loading will not be used!
    bool enabled = 1;
    // File system path (start at root). This folder will contain all the inputs exploded to a separate file.
    string input_path = 2;
    // File system path (start at root). This folder should contain all the outputs for the task as individual files and/or an error text file
    string output_path = 3;
    // In the inputs folder, there will be an additional summary/metadata file that contains references to all files or inlined primitive values.
    LiteralMapFormat format = 4;
    IOStrategy io_strategy = 5;
}

This mechanism allows tasks to be completely portable, using inputs and outputs only from the local file system without any reference to FlyteIDL. However, it's important to note that this feature is currently only supported on Kubernetes.

I hope this helps! If you have any other questions, feel free to ask.

Best,
Dosu

Sources

• Dockerfile.flytepropeller
• flyteidl/protos/flyteidl/core/tasks.proto

---

This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

If you want to continue the conversation, start your reply with @dosu-bot.