Allow attachment of worker to Dead-letter queue

[whambulance]

References Issue #188

Changes aimed at allowing a worker to be assigned as a handler to the dead letter queue. Also publishing the dlq ARN to the yaml file - similar to how the main queue ARN

Reasons:

• It was difficult to add a worker to a DLQ compared to how easy it was to add one to the main queue
• It would have required manually inserting the ARN into the YAML after having created the queue
• Allowing additional manipulation of the DLQ due to exposure of it's ARN in the YAML file

[mnapoli]

Hi, I'm curious why you would want to have a worker on the DLQ?

[whambulance]

Hi @mnapoli - In my specific scenario, I'm using SQS to handle making api calls to a third party provider. If their API is unavailable for some reason, the messages will get pushed into the DLQ, and I need to be able to retry these failed messages at a second provider. This means changing up the content of the message.

Here's another example from the AWS blog, written by some of their Solution Architects.

Essentially, there exists use cases where simply pushing the failed messages back into the main queue isn't viable, and additional processing is required.

[mnapoli]

Thanks for the explanation.

I think that falls outside of the scope of the construct, as it is right now. I don't see that feature being merged at the moment.

Maybe #128 is an alternative that has more chance of happening. Would that work for you?

[whambulance]

As far as the base framework goes, exposing the URL allows you to send messages into the queue, but in order to consume messages from it I believe the ARN needs to be exposed?

Sadly I don't think this would be a viable solution - unless it's modified to expose the DLQ ARN too, similar to what I've done in this PR.

[mnapoli]

Yep, both could be exposed.

[whambulance]

@mnapoli Updated to remove the dlqWorker construct, now only exposes dead letter queue URL & ARN

[mnapoli]

Thanks for the update, that looks good! I have added a few comments inline.

Could you also rebase on master? The tests are now passing on master, and this should allow us to test this PR.

[mnapoli]

That looks good, thanks!

@fredericbarthelet this is a very small addition, could be useful and was reported a few times. WDYT?

[fredericbarthelet]

LGTM :) Merging and issuing release now ;) thanks @whambulance for the contribution !