BP-ICEMAN

As ICEMAN has capability to freeze objects, similarly I'll help to freeze/unfreeze AWS resources by stopping/starting it based on user-defined tags and user-defined actions and increase or descrease the replica count of the K8s resources(Deployment and Statefulset) on the basis of annotation.

SERVICES SUPPORTED

• EC2
• RDS
• k8s
  • Deployment
  • Statefulset

SAMPLE CONF FILE

aws:
  ec2_tags:
    start_stop_schedule : "true"
  region: us-east-2
  rds_tags:
    start_stop_schedule : "true"
  aws_profile: trademo

k8s: 
  context: "internal-trademo.com"
  namespaces: ["default","test"]
  replicas: 3
  deployment_annotations:
    start_stop_schedule : "true"
  sts_annotations:
    start_stop_schedule : "true"

actions_on:
    - k8s:
        - deployment
        - sts
    
    - aws:
        - ec2
        - rds

USAGE

===============================================================

LOCALLY

To run this utility locally from your system.Follow below steps.

• Clone this repo.

• Create your configuration file. Can take reference from config/schedule_resources_sample_config.yml.

• Export two environment variables .

  • CONF_PATH: Path of the configuration file.

  • AWS_SCHEDULE_ACTION: Action which is going to perform i.e start or stop.

  • K8s_SCHEDULE_ACTION: Action which is going to perform i.e resize.

  export CONF_PATH="${pwd}/BP-ICEMAN/config/schedule_resources_sample_config.yml"
  

  export AWS_SCHEDULE_ACTION="start"
  export K8s_SCHEDULE_ACTION="start"
  
  

• Run the python script.

  python3 scripts/schedule_resource_factory.py 
  

USING DOCKER

To run this utility using docker.Follow below steps.

• Clone this repo.

• Make changes to configuration files as required i.e config/schedule_resources.yml.

• Build the docker image using Makefile

   make build VERSION=<Provide image tag here>
  

  

• Run the application

  make run VERSION=<Provide build image tag> CONF_PATH=<Conf-Path> AWS_SCHEDULE_ACTION=<start/stop> K8s_SCHEDULE_ACTION=<start/stop>