Add P&E database credentials, update dependencies, upgrade aws terraform provider

[aloftus23]

Fixes:

• Added PE db creds
• Updated mitmproxy to version 8.0.0 and cryptography to 36.0.2
• Updated terraform config for Terraform AWS Provider Version 4 Error shown below:

Solution (update configurations to their own resources):
hashicorp/terraform-provider-aws#23125 (comment)
https://registry.terraform.io/providers/hashicorp/aws/latest/docs/guides/version-4-upgrade

[epicfaace]

amazing! nice work fixing the terraform 🙌🏽, I had no idea what was happening 😄

[epicfaace]

@aloftus23 see https://stackoverflow.com/questions/61821567/how-do-i-obtain-a-default-value-when-a-tuple-is-empty-in-my-terraform-template

[aloftus23]

@aloftus23 see https://stackoverflow.com/questions/61821567/how-do-i-obtain-a-default-value-when-a-tuple-is-empty-in-my-terraform-template

@epicfaace I added the conditional output for db-accessor so prod doesn't fail. Unfortunately, prevent_destroy doesn't have the functionality to actually ignore a resource replacement; just logs an error.
So I had to take it out. Any issues with the db_accessor resource getting destroyed in prod?

[epicfaace]

So I had to take it out. Any issues with the db_accessor resource getting destroyed in prod?

We should be fine. We should just inspect the output of terraform plan to ensure nothing gets destroyed accidentally. It seems good: https://github.com/cisagov/crossfeed/runs/5851725992?check_suite_focus=true

[epicfaace]

@aloftus23 I think we should uncomment this out, could you do that in a separate PR?