cloud-governance 1.1.315

Cloud Governance Tool

Cloud Governance

What is it?

Cloud Governance tool provides a lightweight and flexible framework for deploying cloud management policies focusing on cost optimize and security. We have implemented several pruning policies.
When monitoring the resources, we found that most of the cost leakage is from available volumes, unused NAT gateways, and unattached Public IPv4 addresses (Starting from February 2024, public IPv4 addresses are chargeable whether they are used or not).

This tool support the following policies: policy

AWS Polices

• Real time Openshift Cluster cost, User cost

• instance_idle: Monitor the idle instances based on the instance metrics for the last 7 days.

  • CPU Percent < 2%
  • Network < 5KiB

• instance_run: List the running ec2 instances.

• unattached_volume: Identify and remove the available EBS volumes.

• zombie_cluster_resource: Identify the non-live cluster resource and delete those resources by resolving dependency. We are deleting more than 20 cluster resources.

  • Ebs, Snapshots, AMI, Load Balancer
  • VPC, Subnets, Route tables, DHCP, Internet Gateway, NatGateway, Network Interface, ElasticIp, Network ACL, Security Group, VPC Endpoint
  • S3
  • IAM User, IAM Role

• ip_unattached: Identify the unattached public IPv4 addresses.

• zombie_snapshots: Identify the snapshots, which are abandoned by the AMI.

• unused_nat_gateway: Identify the unused NatGateway by monitoring the active connection count.

• s3_inactive: Identify the empty s3 buckets, causing the resource quota issues.

• empty_roles: Identify the empty roles that do not have any attached policies to them.

• ebs_in_use: list in use volumes.

• tag_resources: Update cluster and non cluster resource tags fetching from the user tags or from the mandatory tags

• tag_non_cluster: tag ec2 resources (instance, volume, ami, snapshot) by instance name

• tag_iam_user: update the user tags from the csv file

• cost_explorer: Get data from cost explorer and upload to ElasticSearch

• gitleaks: scan GitHub repository git leak (security scan)

• cost_over_usage: send mail to aws user if over usage cost

Azure policies

• instance_idle: Monitor the idle instances based on the instance metrics.
  • CPU Percent < 2%
  • Network < 5KiB
• unattached_volume: Identify and remove the available disks.
• ip_unattached: Identify the unattached public IPv4 addresses.
• unused_nat_gateway: Identify the unused NatGateway by monitoring the active connection count.

IBM policies

• tag_baremetal: Tag IBM baremetal machines
• tag_vm: Tga IBM Virtual Machines machines

** You can write your own policy using Cloud-Custodian and run it (see 'custom cloud custodian policy' in Policy workflows).

Reference:

• The cloud-governance package is placed in PyPi
• The cloud-governance container image is placed in Quay.io
• The cloud-governance readthedocs link is ReadTheDocs

Table of Contents

• Installation
• Configuration
• Run AWS Policy Using Podman
• Run IBM Policy Using Podman
• Run Policy Using Pod
• Pytest
• Post Installation

Installation

Download cloud-governance image from quay.io

# Need to run it with root privileges
sudo podman pull quay.io/cloud-governance/cloud-governance

Environment variables description:

(mandatory)AWS_ACCESS_KEY_ID=$AWS_ACCESS_KEY_ID

(mandatory)AWS_SECRET_ACCESS_KEY=$AWS_SECRET_ACCESS_KEY

Policy name:

(mandatory)policy=instance_idle / instance_run / ebs_unattached / ebs_in_use / tag_cluster_resource / zombie_cluster_resource / tag_ec2_resource

Policy logs output

(mandatory)policy_output=s3://redhat-cloud-governance/logs

Cluster or instance name:

(mandatory policy:tag_cluster_resource)resource_name=ocs-test

Cluster or instance tags:

(mandatory policy:tag_cluster_resource)mandatory_tags="{'Owner': 'Name','Email': '[e-mail]','Purpose': 'test'}"

gitleaks

(mandatory policy: gitleaks)git_access_token=$git_access_token (mandatory policy: gitleaks)git_repo=https://github.com/redhat-performance/cloud-governance (optional policy: gitleaks)several_repos=yes/no (default = no)

Choose a specific region or all for all the regions, default : us-east-2

(optional)AWS_DEFAULT_REGION=us-east-2/all (default = us-east-2)

Choose dry run or not, default yes

(optional)dry_run=yes/no (default = yes)

Choose log level, default INFO

(optional)log_level=INFO (default = INFO)

LDAP hostname to fetch mail records

LDAP_HOST_NAME=ldap.example.com

Enable Google Drive API in console and create Service account

GOOGLE_APPLICATION_CREDENTIALS=$pwd/service_account.json

Configuration

AWS Configuration

Create a user and a bucket

• Create user with IAM
• Create a logs bucket create_bucket.sh

IBM Configuration

• Create classic infrastructure API key

Run AWS Policy Using Podman

# policy=instance_idle
sudo podman run --rm --name cloud-governance -e policy="instance_idle" -e AWS_ACCESS_KEY_ID="$AWS_ACCESS_KEY_ID" -e AWS_SECRET_ACCESS_KEY="$AWS_SECRET_ACCESS_KEY" -e AWS_DEFAULT_REGION="us-east-2" -e dry_run="yes" -e policy_output="s3://bucket/logs" -e log_level="INFO" "quay.io/cloud-governance/cloud-governance"

# policy=instance_run
sudo podman run --rm --name cloud-governance -e policy="instance_run" -e AWS_ACCESS_KEY_ID="$AWS_ACCESS_KEY_ID" -e AWS_SECRET_ACCESS_KEY="$AWS_SECRET_ACCESS_KEY" -e AWS_DEFAULT_REGION="us-east-2" -e dry_run="yes" -e policy_output="s3://bucket/logs" -e log_level="INFO" "quay.io/cloud-governance/cloud-governance"

# select policy ['ec2_stop', 's3_inactive', 'empty_roles', 'ip_unattached', 'unused_nat_gateway', 'zombie_snapshots']
sudo podman run --rm --name cloud-governance -e policy="policy" -e AWS_ACCESS_KEY_ID="$AWS_ACCESS_KEY_ID" -e AWS_SECRET_ACCESS_KEY="$AWS_SECRET_ACCESS_KEY" -e AWS_DEFAULT_REGION="us-east-2" -e dry_run="yes"  -e log_level="INFO" "quay.io/cloud-governance/cloud-governance"

# policy=ebs_unattached
sudo podman run --rm --name cloud-governance -e policy="ebs_unattached" -e AWS_ACCESS_KEY_ID="$AWS_ACCESS_KEY_ID" -e AWS_SECRET_ACCESS_KEY="$AWS_SECRET_ACCESS_KEY" -e AWS_DEFAULT_REGION="us-east-2" -e dry_run="yes" -e policy_output="s3://bucket/logs" -e log_level="INFO" "quay.io/cloud-governance/cloud-governance"

# policy=ebs_in_use
sudo podman run --rm --name cloud-governance -e policy="ebs_in_use" -e AWS_ACCESS_KEY_ID="$AWS_ACCESS_KEY_ID" -e AWS_SECRET_ACCESS_KEY="$AWS_SECRET_ACCESS_KEY" -e AWS_DEFAULT_REGION="us-east-2" -e dry_run="yes" -e policy_output="s3://bucket/logs" -e log_level="INFO" "quay.io/cloud-governance/cloud-governance"

# policy=zombie_cluster_resource
sudo podman run --rm --name cloud-governance -e policy="zombie_cluster_resource" -e AWS_ACCESS_KEY_ID="$AWS_ACCESS_KEY_ID" -e AWS_SECRET_ACCESS_KEY="$AWS_SECRET_ACCESS_KEY" -e AWS_DEFAULT_REGION="us-east-2" -e dry_run="yes" -e resource="zombie_cluster_elastic_ip" -e cluster_tag="kubernetes.io/cluster/test-pd9qq" -e log_level="INFO" "quay.io/cloud-governance/cloud-governance"

# policy=tag_resources
sudo podman run --rm --name cloud-governance -e policy="tag_resources" -e AWS_ACCESS_KEY_ID="$AWS_ACCESS_KEY_ID" -e AWS_SECRET_ACCESS_KEY="$AWS_SECRET_ACCESS_KEY" -e AWS_DEFAULT_REGION="us-east-2" -e tag_operation="read/update/delete" -e mandatory_tags="{'Owner': 'Name','Email': '[e-mail]','Purpose': 'test'}" -e log_level="INFO" -v "/etc/localtime":"/etc/localtime" "quay.io/cloud-governance/cloud-governance"

# policy=tag_non_cluster
sudo podman run --rm --name cloud-governance -e policy="tag_non_cluster" -e AWS_ACCESS_KEY_ID="$AWS_ACCESS_KEY_ID" -e AWS_SECRET_ACCESS_KEY="$AWS_SECRET_ACCESS_KEY" -e AWS_DEFAULT_REGION="us-east-2" -e tag_operation="read/update/delete" -e mandatory_tags="{'Owner': 'Name','Email': '[e-mail]','Purpose': 'test'}" -e log_level="INFO" -v "/etc/localtime":"/etc/localtime" "quay.io/cloud-governance/cloud-governance"

# policy=tag_iam_user
sudo podman run --rm --name cloud-governance -e policy="tag_iam_user" -e AWS_ACCESS_KEY_ID="$AWS_ACCESS_KEY_ID" -e AWS_SECRET_ACCESS_KEY="$AWS_SECRET_ACCESS_KEY" -e user_tag_operation="read/update/delete" -e remove_tags="['Environment', 'Test']" -e username="test_username" -e file_name="tag_user.csv"  -e log_level="INFO" -v "/home/user/tag_user.csv":"/tmp/tag_user.csv" --privileged "quay.io/cloud-governance/cloud-governance"

# policy=cost_explorer
sudo podman run --rm --name cloud-governance -e policy="cost_explorer" -e AWS_ACCESS_KEY_ID="$AWS_ACCESS_KEY_ID" -e AWS_SECRET_ACCESS_KEY="$AWS_SECRET_ACCESS_KEY" -e es_host="$elasticsearch_host" -e es_port="$elasticsearch_port" -e es_index="$elasticsearch_index" -e cost_metric=UnblendedCost -e start_date="$start_date" -e end_date="$end_date" -e granularity="DAILY" -e cost_explorer_tags="['User', 'Budget', 'Project', 'Manager', 'Owner', 'LaunchTime', 'Name', 'Email']" -e log_level="INFO" "quay.io/cloud-governance/cloud-governance:latest"
sudo podman run --rm --name cloud-governance -e policy="cost_explorer" -e AWS_ACCESS_KEY_ID="$AWS_ACCESS_KEY_ID" -e AWS_SECRET_ACCESS_KEY="$AWS_SECRET_ACCESS_KEY" -e es_index="elasticsearch_index" -e cost_metric="UnblendedCost" -e start_date="$start_date" -e end_date="$end_date" -e granularity="DAILY" -e cost_explorer_tags="['User', 'Budget', 'Project', 'Manager', 'Owner', 'LaunchTime', 'Name', 'Email']" -e file_name="cost_explorer.txt" -v "/home/cost_explorer.txt":"/tmp/cost_explorer.txt" -e log_level="INFO" "quay.io/cloud-governance/cloud-governance:latest"

# policy=validate_iam_user_tags
sudo podman run --rm --name cloud-governance  -e policy="validate_iam_user_tags" -e AWS_ACCESS_KEY_ID="$AWS_ACCESS_KEY_ID" -e AWS_SECRET_ACCESS_KEY="$AWS_SECRET_ACCESS_KEY" -e validate_type="spaces/tags" -e user_tags="['Budget', 'User', 'Owner', 'Manager', 'Environment', 'Project']"   -e log_level="INFO" "quay.io/cloud-governance/cloud-governance:latest"

# policy=gitleaks
sudo podman run --rm --name cloud-governance -e policy="gitleaks" -e git_access_token="$git_access_token" -e git_repo="https://github.com/redhat-performance/cloud-governance" -e several_repos="no" -e log_level="INFO" "quay.io/cloud-governance/cloud-governance"

# custom cloud custodian policy (path for custom policy: -v /home/user/custodian_policy:/custodian_policy)
sudo podman run --rm --name cloud-governance -e policy="/custodian_policy/policy.yml" -e AWS_ACCESS_KEY_ID="$AWS_ACCESS_KEY_ID" -e AWS_SECRET_ACCESS_KEY="$AWS_SECRET_ACCESS_KEY" -e AWS_DEFAULT_REGION="us-east-2" -e dry_run="yes" -e policy_output="s3://bucket/logs" -e log_level="INFO" -v "/home/user/custodian_policy":"/custodian_policy" --privileged "quay.io/cloud-governance/cloud-governance"

Run IBM Policy Using Podman

# policy=tag_baremetal
podman run --rm --name cloud-governance -e policy="tag_baremetal" -e account="$account" -e IBM_API_USERNAME="$IBM_API_USERNAME" -e IBM_API_KEY="$IBM_API_KEY" -e SPREADSHEET_ID="$SPREADSHEET_ID" -e GOOGLE_APPLICATION_CREDENTIALS="$GOOGLE_APPLICATION_CREDENTIALS" -v $GOOGLE_APPLICATION_CREDENTIALS:$GOOGLE_APPLICATION_CREDENTIALS -e LDAP_USER_HOST="$LDAP_USER_HOST" -e tag_operation="update" -e log_level="INFO" -v "/etc/localtime":"/etc/localtime" "quay.io/cloud-governance/cloud-governance:latest"

# tag=tab_vm
podman run --rm --name cloud-governance -e policy="tag_vm" -e account="$account" -e IBM_API_USERNAME="$IBM_API_USERNAME" -e IBM_API_KEY="$IBM_API_KEY" -e SPREADSHEET_ID="$SPREADSHEET_ID" -e GOOGLE_APPLICATION_CREDENTIALS="$GOOGLE_APPLICATION_CREDENTIALS" -v $GOOGLE_APPLICATION_CREDENTIALS:$GOOGLE_APPLICATION_CREDENTIALS -e LDAP_USER_HOST="$LDAP_USER_HOST" -e tag_operation="update" -e log_level="INFO" -v "/etc/localtime":"/etc/localtime" "quay.io/cloud-governance/cloud-governance:latest"

Run with yaml config

cp example.yaml env.yaml

Added the supported environment variables. example:

policy: instance_idle
AWS_ACCESS_KEY_ID: ""
AWS_SECRET_ACCESS_KEY: ""

podman run --rm --name cloud-governance \
-v "${PWD}/env.yaml":"/tmp/env.yaml" \
"quay.io/cloud-governance/cloud-governance:latest"

Run Policy Using Pod

Run as a pod job via OpenShift

Job Pod: cloud-governance.yaml

Configmaps: cloud_governance_configmap.yaml

Quay.io Secret: quayio_secret.sh

AWS Secret: cloud_governance_secret.yaml

* Need to convert secret key to base64 [run_base64.py](pod_yaml/run_base64.py)

Pytest

Cloud-governance integration tests using pytest

python3 -m venv governance
source governance/bin/activate
(governance) $ python -m pip install --upgrade pip
(governance) $ pip install coverage
(governance) $ pip install pytest
(governance) $ git clone https://github.com/redhat-performance/cloud-governance
(governance) $ cd cloud-governance
(governance) $ coverage run -m pytest
(governance) $ deactivate
rm -rf *governance*

Post Installation

Delete cloud-governance image

sudo podman rmi quay.io/cloud-governance/cloud-governance