How to debug an Executor


To debug an Executor, you need to look under the hood of Jina. This means things are not straightforward and might change in the future.

Containerized Executor

This How-To does not work for containerized Executors.

It’s very easy to debug Jina Executors. In this How-To, you will learn how to debug Hello Executor step by step.

Step 1: Pull the Executor

Pull the source code of the Executor you want to debug. There are two ways to do that.

jina hub pull jinahub://Hello
from jina import Executor


Step 2: Set the Breakpoints

Go to ~/.jina/hub-package directory. In this folder, there will be one subdirectory for each Executor that you pulled, named by the Executor id. You can find the source files of the Executor in this directory.

Once you locate the source, you can set the breakpoints as you always do.

Step 3: Debug Your Code

Now you can debug your Executor like any Python code. You can either use the Executor on its own or use it inside a Flow.

from jina import Executor

exec = Executor.from_hub('jinahub://Hello')

# Set breakpoint as needed
from docarray import Document
from jina import Flow

f = Flow().add(uses='jinahub://Hello')

with f:
    res ='/', inputs=Document(text='hello'), return_results=True)