tf-state-change

The script performs the following:

1. Downloads the states from s3
2. Copies nlb from state_a to state_b
3. Copies nlb_domain from state_a to state_b
4. Copies nlb_service from state_a to state_b
5. Copies nlb_certificate from state_a to state_b
6. Copies nlb_certificate_validation (route53_record) from state_a to state_b
7. Copies nlb_certificate_validation (acm_certificate_validation) from state_a to state_b
8. Copies nlb_broker_listeners from state_a to state_b keeping target_groups of state_b
9. Copies nlb_service_listeners from state_a to state_b keeping target_groups of state_b
10. Uploads the state of cluster_b to s3

If you set the --prometheus parameter, is performs the following actions additionally:

1. Copies nlb (from prometheus module) from state_a to state_b
2. Copies internal_record_set (from prometheus module) from state_a to state_b
3. Copies nlb_service (from prometheus module) from state_a to state_b
4. Copies nlb_listeners (from prometheus module) from state_a to state_b keeping target_groups of state_b

*You can customize the behaviour in the main function

Install

cd tf-state-change
python3 -m venv venv
. ./venv/bin/activate
pip install -r requirements.txt

Usage

First, get the aws credentials via aws-saml-auth or awsmyid tools so that they are stored in the ~/.aws/credentials

aws-saml-auth
or
awsmyid --login

Usage:

(venv) ab@ANDREIs-MacBook-Pro tf-state-change % python main.py -h
usage: main.py [-h] --states STATES STATES --env ENV [--dry-run] [--prometheus]                                                                                                                                                                                                                                                                                              ─╯

optional arguments:
  -h, --help            show this help message and exit
  --states STATES STATES
                        List of state files to work with
  --env ENV             nonprod or prod
  --dry-run             Run without uploading state back to S3. The changes will be saved to a new "modified" directory
  --prometheus          Set this parameter if prometheus ingress enabled in target clusters

Example:

python main.py --states us-east-1-plygnd-ab-main.tfstate us-east-1-plygnd-ab2-main.tfstate --env nonprod --dry-run
python main.py --states us-east-1-plygnd-ab-main.tfstate us-east-1-plygnd-ab2-main.tfstate --env nonprod

python main.py --states us-east-1-plygnd-ab-main.tfstate us-east-1-plygnd-ab2-main.tfstate --env nonprod --prometheus --dry-run
python main.py --states us-east-1-plygnd-ab-main.tfstate us-east-1-plygnd-ab2-main.tfstate --env nonprod --prometheus

TODO:

2. backup delete resources