cluster_monitoring
Configures an in-cluster monitoring stack for IBM Maximo Application Suite:
- OpenShift user defined project monitoring is enabled (
openshift-monitoring
namespace) - OpenShift monitoring stack is configured to use persistent storage (
openshift-monitoring
namespace) - OpenTelemetry operator is installed (optional,
openshift-operators
namespace) - Grafana installed using the community grafana operator (
grafana
orgrafana5
namespace)
The credentials for the grafana admin user are stored in grafana-admin-credentials
secret in the grafana namespace. A route is created in the grafana namespace to allow access to the grafana UI.
Role Variables
cluster_monitoring_action
Inform the role whether to perform an install
or an uninstall
of the cluster monitoring stack. Can also be set to update_grafana
to update the Grafana Operator from V4 to V5.
Note
When using this role to upgrade from Grafana 4 to 5, the Grafana 5 instance will have a new URL and will not inherit the user database from the old v4 installation, the admin password will be new, and user accounts set up in the v4 instance will need to be recreated in the v5 instance.
- Optional
- Environment Variable:
CLUSTER_MONITORING_ACTION
- Default:
install
cluster_monitoring_include_prometheus
By default this role will reconfigure Prometheus to enable persistent storage and user workload monitoring, this can be disabled by setting this variable to False
.
- Optional
- Environment Variable:
CLUSTER_MONITORING_INCLUDE_PROMETHEUS
- Default:
True
cluster_monitoring_include_grafana
By default Grafana is included in the monitoring stack, this can be disabled by setting this variable to False
.
- Optional
- Environment Variable:
CLUSTER_MONITORING_INCLUDE_GRAFANA
- Default:
True
cluster_monitoring_include_opentelemetry
By default OpenTelemtry is not included in the monitoring stack, this can be enabled by setting this variable to True
.
- Optional
- Environment Variable:
CLUSTER_MONITORING_INCLUDE_OPENTELEMETRY
- Default:
False
Role Variables - Prometheus
prometheus_retention_period
Adjust the retention period for Prometheus metrics, only used when both prometheus_storage_class
and prometheus_alertmgr_storage_class
are set.
- Optional
- Environment Variable:
PROMETHEUS_RETENTION_PERIOD
- Default Value:
15d
prometheus_storage_class
Declare the storage class for Prometheus' metrics data persistent volume.
- Required if one of the known supported storage classes is not installed in the cluster.
- Environment Variable:
PROMETHEUS_STORAGE_CLASS
- Default Value:
ibmc-file-gold-gid
,ocs-storagecluster-cephfs
,azurefiles-premium
(if available)
prometheus_storage_size
Adjust the size of the volume used to store metrics, only used when both prometheus_storage_class
and prometheus_alertmgr_storage_class
are set.
- Optional
- Environment Variable:
PROMETHEUS_STORAGE_SIZE
- Default Value:
20Gi
prometheus_alertmgr_storage_class
Declare the storage class for AlertManager's persistent volume.
- Required if one of the known supported storage classes is not installed in the cluster.
- Environment Variable:
PROMETHEUS_ALERTMGR_STORAGE_CLASS
- Default Value:
ibmc-file-gold-gid
,ocs-storagecluster-cephfs
,azurefiles-premium
(if available)
prometheus_alertmgr_storage_size
Adjust the size of the volume used by AlertManager, only used when both prometheus_storage_class
and prometheus_alertmgr_storage_class
are set.
- Optional
- Environment Variable:
PROMETHEUS_ALERTMGR_STORAGE_SIZE
- Default Value:
20Gi
prometheus_userworkload_retention_period
Adjust the retention period for User Workload Prometheus metrics, this parameter applies only to the User Workload Prometheus instance.
- Optional
- Environment Variable:
PROMETHEUS_USERWORKLOAD_RETENTION_PERIOD
- Default Value:
15d
prometheus_userworkload_storage_class
Declare the storage class for User Workload Prometheus' metrics data persistent volume.
- Optional
- Environment Variable:
PROMETHEUS_USERWORKLOAD_STORAGE_CLASS
- Default Value:
PROMETHEUS_STORAGE_CLASS
prometheus_userworkload_storage_size
Adjust the size of the volume used to store User Workload metrics.
- Optional
- Environment Variable:
PROMETHEUS_USERWORKLOAD_STORAGE_SIZE
- Default Value:
20Gi
Role Variables - Grafana
grafana_major_version
Sets the major version of the grafana operator to install. 4
or 5
- Optional
- Environment Variable:
GRAFANA_MAJOR_VERSION
- Default Value:
5
grafana_v4_namespace
Sets the namespace to install the grafana operator V4 and grafana instance
- Optional
- Environment Variable:
GRAFANA_NAMESPACE
- Default Value:
grafana
grafana_v5_namespace
Sets the namespace to install the grafana operator V5 and grafana instance
- Optional
- Environment Variable:
GRAFANA_V5_NAMESPACE
- Default Value:
grafana5
grafana_instance_storage_class
Declare the storage class for Grafana Instance user data persistent volume.
- Required if one of the known supported storage classes is not installed in the cluster.
- Environment Variable:
GRAFANA_INSTANCE_STORAGE_CLASS
- Default Value:
ibmc-file-gold-gid
,ocs-storagecluster-cephfs
,azurefiles-premium
(if available)
grafana_instance_storage_size
Adjust the size of the volume used to store Grafana user data.
- Optional
- Environment Variable:
GRAFANA_INSTANCE_STORAGE_SIZE
- Default Value:
10Gi
Example Playbook
- hosts: localhost
vars:
prometheus_storage_class: "ibmc-block-gold"
prometheus_alertmgr_storage_class: "ibmc-file-gold-gid"
grafana_instance_storage_class: "ibmc-file-gold-gid"
roles:
- ibm.mas_devops.cluster_monitoring
To Upgrade from Grafana Operator from V4 to V5
- hosts: localhost
vars:
cluster_monitoring_action: "update_grafana"
roles:
- ibm.mas_devops.cluster_monitoring
License
EPL-2.0