With CodeDeploy, Install doesn't end endlessly

In CodeDeploy, sometimes there is no error and the installation does not end endlessly.

If there are no errors, I think that it will be a panic, so it is quite light content, but I will write it as an article.

This happened to me several times,

CodeDeploy's IAM Role doesn't have the required permissions
You are trying to deploy a container that cannot be started due to a mistake in the container configuration file.

In my case, it was either of these two causes.

Case study

For reference, I would like to introduce the situation where the installation that I experienced most recently does not end endlessly.

In Fargate, when I tried to send logs of other containers to firehose using fluent-bit container and FireLens in sidecar configuration, I did not notice the following restrictions,

If you use a custom configuration file, you must specify a different path than the one used by FireLens. Amazon ECS reserves file paths for /fluent-bit/etc/fluent-bit.conf (Fluent Bit) and /fluentd/etc/fluent.conf (Fluentd).

FROM amazon/aws-for-fluent-bit:2.1.1

COPY conf/fluent-bit/fluent-bit.conf /fluent-bit/etc/fluent-bit.conf

Because I was trying to deploy the image created from the Dockerfile that had the configuration file in the reserved path like this

This config file was due to the fact that it was a custom config file and wasn't enough for a regular config file.

This may be the reason why the installation did not finish.

As a countermeasure,

--Check if the container works locally --If you're running locally, review your IAM surroundings

Is it about?

Since there are no errors, I think it will be quite impatient if I fall into this phenomenon for the first time, but I hope that this article will lead to a general examination and lead to a solution.