Synchronizes groups from external providers into OpenShift
The OpenShift Container Platform contains functionality to synchronize groups found in external identity providers into the platform. Currently, the functionality that is included in OpenShift is limited to synchronizing LDAP only. This operator is designed to integrate with external providers in order to provide new solutions.
Group Synchronization is facilitated by creating a GroupSync
resource. The following describes the high level schema for this resource:
apiVersion: redhatcop.redhat.io/v1alpha1
kind: GroupSync
metadata:
name: example-groupsync
spec:
providers:
- <One or more providers to synchronize>
Use the following steps to deploy the operator to an OpenShift cluster
git clone https://github.com/redhat-cop/group-sync-operator.git
cd group-sync-operator
make deploy IMG=quay.io/redhat-cop/group-sync-operator:latest
Note: The make deploy
command will execute the manifests
target that will require additional build tools to be made available. This target can be skipped by including the -o manifests
in the command above.
In most cases, authentication details must be provided in order to communicate with providers. Authentication details are provider specific with regards to the required values. In supported providers, the secret can be referenced in the credentialsSecret
by name and namespace where it has been created as shown below:
credentialsSecret:
name: <secret_name>
namespace: <secret_namespace>
Integration with external systems is made possible through a set of pluggable external providers. The following providers are currently supported:
The following sections describe the configuration options available for each provider
Groups contained within Azure Active Directory can be synchronized into OpenShift. The following table describes the set of configuration options for the Azure provider:
Name | Description | Defaults | Required |
---|---|---|---|
authorityHost |
Azure Active Directory Endpoint | https://login.microsoftonline.com |
No |
baseGroups |
List of groups to start searching from instead of listing all groups in the directory | No | |
credentialsSecret |
Name of the secret containing authentication details (See below) | Yes | |
filter |
Graph API filter | No | |
groups |
List of groups to filter against | No | |
userNameAttributes |
Fields on a user record to use as the User Name | userPrincipalName |
No |
prune |
Prune Whether to prune groups that are no longer in Azure | false |
No |
The following is an example of a minimal configuration that can be applied to integrate with a Azure provider:
apiVersion: redhatcop.redhat.io/v1alpha1
kind: GroupSync
metadata:
name: azure-groupsync
spec:
providers:
- name: azure
azure:
credentialsSecret:
name: azure-group-sync
namespace: group-sync-operator
Authentication to Azure can be performed using Application Registration with access to query group information in Azure Active Directory.
The App Registration must be granted access to the following Microsoft Graph API's:
A secret must be created in the same namespace that contains the GroupSync
resource:
The following keys must be defined in the secret
AZURE_TENANT_ID
- Tenant IDAZURE_CLIENT_ID
- Client IDAZURE_CLIENT_SECRET
- Client SecretThe secret can be created by executing the following command:
oc create secret generic azure-group-sync --from-literal=AZURE_TENANT_ID=<AZURE_TENANT_ID> --from-literal=AZURE_CLIENT_ID=<AZURE_CLIENT_ID> --from-literal=AZURE_CLIENT_SECRET=<AZURE_CLIENT_SECRET>
Teams stored within a GitHub organization can be synchronized into OpenShift. The following table describes the set of configuration options for the GitHub provider:
Name | Description | Defaults | Required |
---|---|---|---|
ca |
Reference to a resource containing a SSL certificate to use for communication (See below) | No | |
caSecret |
DEPRECATED Reference to a secret containing a SSL certificate to use for communication (See below) | No | |
credentialsSecret |
Reference to a secret containing authentication details (See below) | Yes | |
insecure |
Ignore SSL verification | false |
No |
organization |
Organization to synchronize against | Yes | |
teams |
List of teams to filter against | No | |
url |
Base URL for the GitHub or GitHub Enterprise host (Must contain a trailing slash) | No | |
prune |
Prune Whether to prune groups that are no longer in GitHub | false |
No |
The following is an example of a minimal configuration that can be applied to integrate with a GitHub provider:
apiVersion: redhatcop.redhat.io/v1alpha1
kind: GroupSync
metadata:
name: github-groupsync
spec:
providers:
- name: github
github:
organization: ocp
credentialsSecret:
name: github-group-sync
namespace: group-sync-operator
Authentication to GitHub can be performed using an OAuth Personal Access Token or as an GitHub App, using a secret key and appId.
The OAuth Personal Access Token needs to scope of admin:org/read:org
.
A secret must be created in the same namespace that contains the GroupSync
resource:
When using an OAuth token, the following key is required:
token
- OAuth tokenThe secret can be created by executing the following command:
oc create secret generic github-group-sync --from-literal=token=<token>
When authenticating as a Github App, the following keys are required:
privateKey
and appId
In GitHub, go to developer-settings -> github apps.
mapByScimId
, this permissions needs to be Read & Write, though the operator only does read-only operations. The reason for this is the use of the v4 graphql api-endpoint.The secret can be created by executing the following command:
oc create secret generic github-group-sync --from-literal=appId=<theAppId> --from-file=privateKey=</path/to/thefile>
Groups stored within a GitLab can be synchronized into OpenShift. The following table describes the set of configuration options for the GitLab provider:
Name | Description | Defaults | Required |
---|---|---|---|
ca |
Reference to a resource containing a SSL certificate to use for communication (See below) | No | |
caSecret |
DEPRECATED Reference to a secret containing a SSL certificate to use for communication (See below) | No | |
credentialsSecret |
Reference to a secret containing authentication details (See below) | Yes | |
insecure |
Ignore SSL verification | false |
No |
groups |
List of groups to filter against | No | |
prune |
Prune Whether to prune groups that are no longer in GitLab | false |
No |
scope |
Scope for group synchronization. Options are one for one level or sub to include subgroups |
sub |
No |
url |
Base URL for the GitLab instance | https://gitlab.com |
No |
The following is an example of a minimal configuration that can be applied to integrate with a GitHub provider:
apiVersion: redhatcop.redhat.io/v1alpha1
kind: GroupSync
metadata:
name: gitlab-groupsync
spec:
providers:
- name: gitlab
gitlab:
credentialsSecret:
name: gitlab-group-sync
namespace: group-sync-operator
Authentication to GitLab can be performed using a Token or a Username and Password (Note: 2FA not supported). A secret must be created in the same namespace that contains the GroupSync
resource:
When using an OAuth token, the following token types are supported:
Personal Access Token
OAuth Token
Job Token
the following key is required:
token
- OAuth token
Optionally, the tokenType
key can be specified to indicate the type of token being provided from the following values:
oauth
personal
job
If no tokenType
is provided, oauth
is used by default
The secret can be created by executing the following command:
oc create secret generic gitlab-group-sync --from-literal=token=<token>
To specify a token type, such as a Personal Access Token, the following command can be executed:
oc create secret generic gitlab-group-sync --from-literal=token=<token> --from-literal=tokenType=personal
The following keys are required for username and password:
username
- Username for authenticating with GitLabpassword
- Password for authenticating with GitLabThe secret can be created by executing the following command:
oc create secret generic gitlab-group-sync --from-literal=username=<username> --from-literal=password=<password>
Groups stored within an LDAP server can be synchronized into OpenShift. The LDAP provider implements the included features of the Syncing LDAP groups feature and makes use of the libraries from the OpenShift Command Line tool to streamline the migration to this operator based implementation.
The configurations of the three primary schemas (rfc2307
, activeDirectory
and augmentedActiveDirectory
) can be directly migrated as is without any modification.
Name | Description | Defaults | Required |
---|---|---|---|
ca |
Reference to a resource containing a SSL certificate to use for communication (See below) | No | |
caSecret |
DEPRECATED Reference to a secret containing a SSL certificate to use for communication (See below) | No | |
credentialsSecret |
Reference to a secret containing authentication details (See below) | No | |
insecure |
Ignore SSL verification | false |
No |
groupUIDNameMapping |
User defined name mapping | No | |
rfc2307 |
Configuration using the rfc2307 schema | No | |
activeDirectory |
Configuration using the activeDirectory schema | No | |
augmentedActiveDirectory |
Configuration using the activeDirectory schema | No | |
url |
Connection URL for the LDAP server | ldap://ldapserver:389 |
No |
whitelist |
Explicit list of groups to synchronize | No | |
blacklist |
Explicit list of groups to not synchronize | No | |
prune |
Prune Whether to prune groups that are no longer in LDAP | false |
No |
The following is an example using the rfc2307
schema:
apiVersion: redhatcop.redhat.io/v1alpha1
kind: GroupSync
metadata:
name: ldap-groupsync
spec:
providers:
- ldap:
credentialsSecret:
name: ldap-group-sync
namespace: group-sync-operator
insecure: true
rfc2307:
groupMembershipAttributes:
- member
groupNameAttributes:
- cn
groupUIDAttribute: dn
groupsQuery:
baseDN: ou=Groups,dc=example,dc=com
derefAliases: never
filter: (objectClass=groupofnames)
scope: sub
tolerateMemberNotFoundErrors: true
tolerateMemberOutOfScopeErrors: true
userNameAttributes:
- cn
userUIDAttribute: dn
usersQuery:
baseDN: ou=Users,dc=example,dc=com
derefAliases: never
scope: sub
url: ldap://ldapserver:389
name: ldap
The examples provided in the OpenShift documented referenced previously can be used to construct the schemas for the other LDAP synchronization types.
If authentication is required in order to communicate with the LDAP server, a secret should be created in the same namespace that contains the GroupSync
resource. The following keys can be defined:
username
- Username (Bind DN) for authenticating with the LDAP serverpassword
- Password for authenticating with the LDAP serverThe secret can be created by executing the following command:
oc create secret generic ldap-group-sync --from-literal=username=<username> --from-literal=password=<password>
Groups can be explicitly whitelisted or blacklisted in order to control the groups that are eligible to be synchronized into OpenShift. When running LDAP group synchronization using the command line, this configuration is referenced via separate files, but these are instead specified in the blacklist
and whitelist
properties as shown below:
apiVersion: redhatcop.redhat.io/v1alpha1
kind: GroupSync
metadata:
name: ldap-groupsync
spec:
providers:
- ldap:
...
whitelist:
- cn=Online Corporate Banking,ou=Groups,dc=example,dc=com
...
name: ldap
apiVersion: redhatcop.redhat.io/v1alpha1
kind: GroupSync
metadata:
name: ldap-groupsync
spec:
providers:
- ldap:
...
blacklist:
- cn=Finance,ou=Groups,dc=example,dc=com
...
name: ldap
Groups stored within Keycloak can be synchronized into OpenShift. The following table describes the set of configuration options for the Keycloak provider:
Name | Description | Defaults | Required |
---|---|---|---|
ca |
Reference to a resource containing a SSL certificate to use for communication (See below) | No | |
caSecret |
DEPRECATED Reference to a secret containing a SSL certificate to use for communication (See below) | No | |
credentialsSecret |
Reference to a secret containing authentication details (See below) | Yes | |
groups |
List of groups to filter against | No | |
insecure |
Ignore SSL verification | false |
No |
loginRealm |
Realm to authenticate against | master |
No |
realm |
Realm to synchronize | Yes | |
scope |
Scope for group synchronization. Options are one for one level or sub to include subgroups |
sub |
No |
url |
Base URL for the Keycloak server. Older versions (<17.0.0) including Red Hat SSO should include the context path /auth appended to the hostname |
Yes | |
prune |
Prune Whether to prune groups that are no longer in Keycloak | false |
No |
The following is an example of a minimal configuration that can be applied to integrate with a Keycloak provider:
apiVersion: redhatcop.redhat.io/v1alpha1
kind: GroupSync
metadata:
name: keycloak-groupsync
spec:
providers:
- name: keycloak
keycloak:
realm: ocp
credentialsSecret:
name: keycloak-group-sync
namespace: group-sync-operator
url: https://keycloak-keycloak-operator.apps.openshift.com/auth
A user with rights to query for Keycloak groups must be available. The following permissions must be associated to the user:
A secret must be created in the same namespace that contains the GroupSync
resource. It must contain the following keys for the user previously created:
username
- Username for authenticating with Keycloakpassword
- Password for authenticating with KeycloakThe secret can be created by executing the following command:
oc create secret generic keycloak-group-sync --from-literal=username=<username> --from-literal=password=<password>
Okta Groups assigned to Okta Applications can be synchronized into OpenShift. The developer docs for the Okta API that the Okta Syncer uses can be found here. The following table describes the set of configuration options for the Okta provider:
Name | Description | Defaults | Required |
---|---|---|---|
credentialsSecret |
Reference to a secret containing authentication details (See below) | '' |
Yes |
groups |
List of groups to filter against | nil |
No |
url |
Okta URL which can be found under the "Okta Domain" in your application settings (must contain the scheme and a trailing slash) | '' |
Yes |
appId |
Okta Application (Client) ID that is attached to the application groups you wish to sync | '' |
Yes |
extractLoginUsername |
Bool to determine if you should extract username from okta login | false |
No |
profileKey |
Attribute field on Okta User Profile you would like to use as identity | 'login' |
No |
groupLimit |
Integer to set the maximum number of groups to retrieve from OKTA per request. | 1000 |
No |
prune |
Prune Whether to prune groups that are no longer in OKTA | false |
No |
The following is an example of a minimal configuration that can be applied to integrate with an Okta provider:
apiVersion: redhatcop.redhat.io/v1alpha1
kind: GroupSync
metadata:
name: okta-sync
spec:
providers:
- name: okta
okta:
credentialsSecret:
name: okta-api-token
namespace: group-sync-operator
url: "https://example.okta.com/"
appId: xxxxxxxxxxxxxxxxxxxx
A secret must be created in the same namespace as the group-sync-operator pod. It must contain the following key:
okta-api-token
- Okta API Token for interacting with OktaThe secret can be created by executing the following command:
oc create secret generic okta-api-token --from-literal=okta-api-token=<OKTA_API_TOKEN> -n group-sync-operator
Additional metadata based on Keycloak group are also added to the OpenShift groups as Annotations including:
Several providers allow for certificates to be provided in either a ConfigMap or Secret to communicate securely to the target host through the use of a property called ca
.
The certificate can be added to a Secret called keycloak-certs using the key ca.crt
representing the certificate using the following command.
oc create secret generic keycloak-certs --from-file=ca.crt=<file>
An example of how the CA certificate can be added to the Keycloak provider is shown below:
apiVersion: redhatcop.redhat.io/v1alpha1
kind: GroupSync
metadata:
name: keycloak-groupsync
spec:
providers:
- name: keycloak
keycloak:
realm: ocp
credentialsSecret:
name: keycloak-group-sync
namespace: group-sync-operator
ca:
kind: Secret
name: keycloak-certs
namespace: group-sync-operator
key: ca.crt
url: https://keycloak-keycloak-operator.apps.openshift.com
Alteratively, a ConfigMap can be used instead instead of a Secret. This is useful when using the Certificate injection using Operators feature.
The following command can be used to create a ConfigMap containing the certificate:
oc create configmap keycloak-certs --from-file=ca.crt=<file>
An example of how the CA certificate can be added to the Keycloak provider is shown below:
apiVersion: redhatcop.redhat.io/v1alpha1
kind: GroupSync
metadata:
name: keycloak-groupsync
spec:
providers:
- name: keycloak
keycloak:
realm: ocp
credentialsSecret:
name: keycloak-group-sync
namespace: group-sync-operator
ca:
kind: ConfigMap
name: keycloak-certs
namespace: group-sync-operator
key: ca.crt
url: https://keycloak-keycloak-operator.apps.openshift.com
A cron style expression can be specified for which a synchronization event will occur. The following specifies that a synchronization should occur nightly at 3AM
apiVersion: redhatcop.redhat.io/v1alpha1
kind: GroupSync
metadata:
name: keycloak-groupsync
spec:
schedule: "0 3 * * *"
providers:
- ...
If a schedule is not provided, synchronization will occur only when the object is reconciled by the platform.
By default, the operator monitors resources in the namespace that it has been deployed within. This is defined by setting the WATCH_NAMESPACE
environment variable. Support is available for accessing ConfigMaps and Secrets in other namespaces so that existing resources may be utilized as desired.
To enable the operator to access resources across multiple, set the environment variable with a comma separate list of namespaces that include the namespace the operator is deployed within and any additional namespaces that are desired.
To make use of this feature when deploying through the Operator Lifecycle Manager, set the following configuration on the Subscription
resource:
apiVersion: operators.coreos.com/v1alpha1
kind: Subscription
metadata:
name: group-sync-operator
namespace: group-sync-operator
spec:
channel: alpha
installPlanApproval: Automatic
name: group-sync-operator
source: community-operators
sourceNamespace: openshift-marketplace
config:
env:
- name: WATCH_NAMESPACE
value: "<comma separated list of namespaces>"
This is a namespace level operator that you can deploy in any namespace. However, group-sync-operator
is recommended.
It is recommended to deploy this operator via OperatorHub
, but you can also deploy it using Helm
.
If you want to utilize the Operator Lifecycle Manager (OLM) to install this operator, you can do so in two ways: from the UI or the CLI.
Arch | Support |
---|---|
amd64 | ✅ |
arm64 | ✅ |
ppc64le | ✅ |
s390x | ✅ |
group sync operator
. This will then return an item for our operator and you can select it to get started. Once you've arrived here, you'll be presented with an option to install, which will begin the process.group-sync-operator
is presented and can be created automatically at installation time.Automatic
or Manual
).Subscribe
and the installation will begin. After a few moments you can go ahead and check your namespace and you should see the operator running.If you'd like to launch this operator from the command line, you can use the manifests contained in this repository by running the following:
oc new-project group-sync-operator
oc apply -f config/operatorhub -n group-sync-operator
This will create the appropriate OperatorGroup and Subscription and will trigger OLM to launch the operator in the specified namespace.
Here are the instructions to install the latest release with Helm.
oc new-project group-sync-operator
helm repo add group-sync-operator https://redhat-cop.github.io/group-sync-operator
helm repo update
helm install group-sync-operator group-sync-operator/group-sync-operator
This can later be updated with the following commands:
helm repo update
helm upgrade group-sync-operator group-sync-operator/group-sync-operator
Prometheus compatible metrics are exposed by the Operator and can be integrated into OpenShift's default cluster monitoring. To enable OpenShift cluster monitoring, label the namespace the operator is deployed in with the label openshift.io/cluster-monitoring="true"
.
oc label namespace <namespace> openshift.io/cluster-monitoring="true"
export operatorNamespace=group-sync-operator-local # or group-sync-operator
oc label namespace ${operatorNamespace} openshift.io/cluster-monitoring="true"
oc rsh -n openshift-monitoring -c prometheus prometheus-k8s-0 /bin/bash
export operatorNamespace=group-sync-operator-local # or group-sync-operator
curl -v -s -k -H "Authorization: Bearer $(cat /var/run/secrets/kubernetes.io/serviceaccount/token)" https://group-sync-operator-controller-manager-metrics-service.${operatorNamespace}.svc.cluster.local:8443/metrics
exit
make install
export repo=redhatcopuser #replace with yours
docker login quay.io/$repo/group-sync-operator
make docker-build IMG=quay.io/$repo/group-sync-operator:latest
make docker-push IMG=quay.io/$repo/group-sync-operator:latest
oc new-project group-sync-operator-local
kustomize build ./config/local-development | oc apply -f - -n group-sync-operator-local
export token=$(oc serviceaccounts get-token 'group-sync-operator-controller-manager' -n group-sync-operator-local)
oc login --token ${token}
make run ENABLE_WEBHOOKS=false
Define an image and tag. For example...
export imageRepository="quay.io/redhat-cop/group-sync-operator"
export imageTag="$(git -c 'versionsort.suffix=-' ls-remote --exit-code --refs --sort='version:refname' --tags https://github.com/redhat-cop/group-sync-operator.git '*.*.*' | tail --lines=1 | cut --delimiter='/' --fields=3)"
Deploy chart...
make helmchart IMG=${imageRepository} VERSION=${imageTag}
helm upgrade -i group-sync-operator-local charts/group-sync-operator -n group-sync-operator-local --create-namespace
Delete...
helm delete group-sync-operator-local -n group-sync-operator-local
kubectl delete -f charts/group-sync-operator/crds/crds.yaml
export repo=redhatcopuser #replace with yours
docker login quay.io/$repo/group-sync-operator
make docker-build IMG=quay.io/$repo/group-sync-operator:latest
make docker-push IMG=quay.io/$repo/group-sync-operator:latest
make manifests
make bundle IMG=quay.io/$repo/group-sync-operator:latest
operator-sdk bundle validate ./bundle --select-optional name=operatorhub
make bundle-build BUNDLE_IMG=quay.io/$repo/group-sync-operator-bundle:latest
docker login quay.io/$repo/group-sync-operator-bundle
docker push quay.io/$repo/group-sync-operator-bundle:latest
operator-sdk bundle validate quay.io/$repo/group-sync-operator-bundle:latest --select-optional name=operatorhub
oc new-project group-sync-operator
oc label namespace group-sync-operator openshift.io/cluster-monitoring="true"
operator-sdk cleanup group-sync-operator -n group-sync-operator
operator-sdk run bundle -n group-sync-operator quay.io/$repo/group-sync-operator-bundle:latest
git tag -a "<tagname>" -m "<commit message>"
git push upstream <tagname>
If you need to remove a release:
git tag -d <tagname>
git push upstream --delete <tagname>
If you need to "move" a release to the current master
git tag -f <tagname>
git push upstream -f <tagname>
operator-sdk cleanup group-sync-operator -n group-sync-operator
oc delete operatorgroup operator-sdk-og
oc delete catalogsource group-sync-operator-catalog