eloq operator Helm Chart

The current chart is for installing the eloq-operator (including CRD for EloqDBCluster) and the monitoring infrastructure used by EloqDBCluster.

NOTE: The monitoring infrastructure is optional and depends on whether it is already installed in your current environment. They come from the prometheus and grafana communities: prometheus, grafana, lock-distributed, mimir.

Install the eloq-operator

Helm must be installed to use the charts. Please refer to Helm's documentation to get started.

Once Helm has been set up correctly, add the repo as follows:

helm repo add eloqdata https://monographdb.github.io/monograph-charts/
helm repo update
# for example: helm install eloq-operator eloqdata/eloq-operator --namespace eloq-operator-system
helm install [RELEASE_NAME]  monographdb/eloq-operator --namespace [NAMESPACE_NAME]

NOTE: If the installation specifies namespace please create it first.

Check the installed eloq operator

# for example: helm list --namespace eloq-operator-system / helm list --all --all-namespaces
helm list --namespace [NAMESPACE_NAME]

Uninstall Helm Release and CRDs

The default value of keepCrds in values.yaml is set to true. This means that even after uninstalling the release, Custom Resource Definitions (CRDs) associated with the release are retained. This behavior is particularly useful for preserving CRDs that you might want to keep for future use or to maintain data integrity.

To uninstall a Helm release, use the following command:

helm uninstall [RELEASE_NAME] --namespace [NAMESPACE_NAME]

In cases where you wish to delete the CRDs manually, use the following command:

kubectl delete crd EloqDBClusters.monograph-service.monographdb.com

Alternatively, setting keepCrds to false will result in the automatic deletion of the associated CRDs when the Helm release is uninstalled.

eloq-operator chart arguments

The following parameters can be overridden by helm --set. For example: --set controllerManager.serviceAccoun=" eloq-op-sa"

Name	Type	Default Value	Description
nameOverride	string	""	Overrides the "eloq-operator" with this name.
controllerManager.serviceAccount.name	string	eloq-operator-controller-manager-sa	The service account name of the eloq operator controller manager pods.
controllerManager.serviceAccount.annotations	object	{}	Annotations for the controllerManager.serviceAccount.
controllerManager.image.repository	string	monographdb/eloq-operator	The image name of the eloq operator.
controllerManager.image.tag	string	v1.0.4	The version tag for eloq operator docker image.
controllerManager.imagePullPolicy	string	IfNotPresent	-
controllerManager.imagePullSecrets	object	{}	-
controllerManager.resources	object	Same format as k8s resource	Resource requests and limits for eloq operator controller manager pods.
controllerManager.healthPort	string	18080	-
controllerManager.metricPort	string	18081	-
controllerManager.watchNamespaces	string	"" (watch all namespaces)	Set the controller to watch specific namespaces instead of all. (e.g. "", "NAMESPACE", or "N1,N2,N3")
controllerManager.serviceMonitor.release	string	kube-prometheus-stack	Set the release name for the controller's metric service monitor.
controllerManager.env	list	[]	The environment variable of the eloq operator controller manager pods.
keepCrds	bool	true	Keep or not keep CRDs when uninstalling the helm release.
cert-manager.enabled	bool	false	Set certManager.enabled=true will install the cert-menager to release.namespace.

Install monitoring infrastructure

Currently, eloq-monitoring only supports AWS. It requires you to create a serviceAccount before execution so that it can access AWS resources outside the EKS cluster. This includes S3.

# for example: helm install eloq-monitor-stack monographdb/eloq-monitoring --namespace eloq-monitoring-ns
helm install [RELEASE_NAME] monographdb/eloq-monitoring --namespace [NAMESPACE]

eloq-monitoring chart arguments

Name	Type	Default value	Description
global.namespace	string	eloq-monitoring-ns	Monitor the namespace where the component is installed.
global.serviceAccount.name	string	eloq-monitoring-sa	ServiceAccount
aws.region	string	ap-northeast-1	AWS configuration.
aws.s3.endpoint	string	s3.ap-northeast-1.amazonaws.com	S3 endpoint.
kube-prometheus-stack.enable	bool	true	If or not kube-prometheus-stack is installed.
kube-prometheus-stack.storageSpec.storageClassName	string	ebs-sc-gp3
kube-prometheus-stack.storageSpec.storageSize	string	128Gi
loki-distributed.enable	bool	true	If or not loki-distributed is installed.
loki-distributed.bucketnames	string	eloq-loki-data
loki-distributed.ingester.persistence.walStorageSpec.size	string	12Gi
loki-distributed.ingester.persistence.dataStorageSpec.size	string	12Gi
loki-distributed.ingester.persistence.walStorageSpec.storageClass	string	ebs-sc-gp3
loki-distributed.ingester.persistence.dataStorageSpec.storageClass	string	ebs-sc-gp3
mimir-distributed.enabled	bool	true
mimir-distributed.ingester.persistence.dataStorageSpec.storageClass	string	ebs-sc-gp3
mimir-distributed.storageClass	string	ebs-sc-gp3
mimir-distributed.ruler_storage.bucket_name	string	eloq-mimir-ruler	S3 bucketname of ruler_storage
mimir-distributed.alertmanager_storage.bucket_name	string	eloq-mimir-alertmanager	S3 bucketname of alertmanager_storage
mimir-distributed.blocks_storage.bucket_name	string	eloq-mimir-data	S3 bucketname of blocks_storage
promtail.enable	bool	true	If or not promtail is installed.

NOTE: All of the above S3 bucketname need to already exist.