6. Container interaction and debugging

Container interaction

Sometimes things don’t go as smoothly as expected.

For such rare cases, Dagger gives us the possibility to start an interactive session to inspect a specific container:

dagger call backend --context=. terminal

This attaches us to sh shell inside the container returned by the bakend function. To exit press ctrl-D repeatedly.

If you prefer another command, you can pass the optional cmd argument:

dagger call backend --context=. terminal --cmd=printenv

In this case it executes the command and terminates afterward.

Task 6.1: Start a live terminal session

Connect to the frontend container with a live terminal session using bash and list the installed debian packages.

dagger call frontend --context=frontend terminal --cmd=bash
# dpkg -l

Debugging

Debug flag

In other cases, it would be helpful if we just could get a bit more detailed output.

For this purpose, every Dagger command can be invoked with the debug flag:

dagger call --debug backend --context=.

Interactive flag

Dagger has an interactive debugging feature, which allows users to drop in to an interactive shell when a pipeline run fails, with all the context at the point of failure.

No need to set breakpoints or change your code.

Just do your dagger call and include the --interactive flag. E.g:

dagger call --interactive foo

Use the interactive (-i) shorthand to enter the container when a problem occurs:

dagger -i -c 'container | from alpine | with-exec echooo "Daggernaut" | stdout'

Utilizing the interactive flag, you can also set breakpoints in your pipeline .terminal()

import dagger
from dagger import dag, function, object_type


@object_type
class MyModule:
    @function
    async def foo(self) -> dagger.Container:
        return await (
            dag.container()
            .from_("alpine:latest")
            .terminal()
            .with_exec(["sh", "-c", "echo hello world > /foo"])
            .terminal()
        )

Debugging with Dagger Cloud

With https://dagger.io/cloud[Dagger Cloud] you gain visibility into every aspect of your Dagger pipelines. It gives you insights across all pipelines, both pre-push and post-push.

As a single user you can register for free https://dagger.cloud/signup[here] .

The https://docs.dagger.io/configuration/cloud/[Dagger Cloud documentation] shows how to connect to https://dagger.cloud/ to see your pipeline traces.

When you are connected and your local pipeline runs, just type w. This will open a browser showing the traces ot this run.

Give it a try :)