To be used with the AWS CDK for Python
CrossAccountDomainDelegation
Creates delegated domains
from ca_cdk_constructs import CrossAccountDomainDelegation, R53ParentZoneConfig
zone = PublicHostedZone(self, "Zone", zone_name="my-subdomain.acme.org")
# creates delegation records in for my-subdomain.acme.org in acme.org
CrossAccountDomainDelegation(
subdomain_stack,
"MySubdomainDnsDelegation",
parent_zone_config=R53ParentZoneConfig(
account_id="1234566789012",
zone_name="acme.org",
role_name="R53UpdateRole" # existing role in the parent zone account
),
hosted_zone=zone
)
WAF Rule Templates
TODO TODO TODO
ca_cdk_constructs
includes a vendored cdk8s library that is compatible with the currently supported CA Kubernetes platform version.
You need to update the version of ca_cdk_constructs
in your project if both of these are correct:
- your project deploys k8s resources using the vendored cdk8s library
- the CA platform was updated to a new version
To vendor the libraries for the K8s version currently supported by the CA Kubernetes platform:
- update the k8s / crd versions in
cdk8s.yaml
- run
cdk8s import --output ca_cdk_constructs/eks/imports
The library can be then used as follows:
import ca_cdk_constructs.eks.imports.k8s as k8s
k8s.KubeDeployment(.....)
EksClusterIntegration
Makes it possible to deploy to imported EKS clusters.
from ca_cdk_constructs.eks import EksClusterIntegration
# in an existing stack
eks_integration = EksClusterIntegration(self, "EksIntegration", vpc=vpc, cluster_name="mycluster")
# for imported clusters the kubectl role must be manually added to aws-auth
# The role ARN will also be available in the K8sAuthRoleArn output
eks_integration.role
# the EKS cluster
eks_integration.cluster
ExternalAwsSecretsChart
cdk8s Chart to deploy External Secrets referencing one or more AWS SecretsManager or ParameterStore secrets.
See the tests
ModifyDBClusterPassword
Modifies the password of an Aurora cluster
modify_cluster_password = ModifyDBClusterPassword(self, "ModifyClusterPassword", cluster_id=cluster_id, secret=db_secret)
modify_cluster_password.trigger_on_stack_create_update()
# access the udnerlaying lambda to e.g. add it to a state machine
modify_cluster_password.lambda_funct
AuroraFastClone
Clones an Aurora cluster.
from ca_cdk_constructs.storage.aurora_clone_refresh import AuroraCloneRefresh
source_cluster = DatabaseCluster(self, "AuroraCluster", ....) # or lookup one
vpc = source_cluster.vpc # or look it up
cluster_pg = CfnDBClusterParameterGroup(
self,
"DBClusterParameterGroup",
description=f"Cluster parameter group for test clone",
family=source_cluster.engine.parameter_group_family,
parameters={"log_hostname": 1},
)
cluster_instance_pg = rds.CfnDBParameterGroup(
self,
"DBParameterGroup",
description=f"DB parameter group for test clone instance",
family=source_cluster.engine.parameter_group_family,
parameters={"log_hostname": 1},
)
# periodically clone the source cluster
cloned_cluster = AuroraCloneRefresh(self, "TestClone",
source_cluster=source_cluster,
source_cluster_vpc=vpc,
source_cluster_master_username=username,
db_instance_class="db.t3.medium",
cluster_parameter_group=cluster_pg,
instance_parameter_group=cluster_instance_pg,
tags={
"Tag": "Value"
},
clone_schedule=Schedule.cron(minute="0", hour="8"),
notifications_topic=topic)
# allow access to the clone from certain ranges
cloned_cluster.allow_from(ec2.Peer.ipv4(vpc.vpc_cidr_block))
# or
clone.cluster_sg.allow_....
# access the cloned cluster credentials
cloned_cluster.clone_secret # DatabaseSecret
# the clone SNS topic
cloned_cluster.notifications_topic # Topic
# the event rule
cloned_cluster.event_rule
poetry install
poetry run pytest
To run tests against a particular version of python (e.g. 3.10), run poetry env use 3.10
before poetry install
. You do need to have that version installed on your system through brew install [email protected]
first.
Versioning and depoyments are done via Github releases. Information on creating releases can be found here. Please follow the existing convention for version names and use automatic patch notes for consistence unless a major change is being made.
When making a new release, please ensure that the package version in pyproject.toml
is updated with the version matching the tag name.