diff options
Diffstat (limited to 'roles/openshift_metrics/README.md')
| -rw-r--r-- | roles/openshift_metrics/README.md | 80 |
1 files changed, 62 insertions, 18 deletions
diff --git a/roles/openshift_metrics/README.md b/roles/openshift_metrics/README.md index 30a0a608d..f4c47c7bb 100644 --- a/roles/openshift_metrics/README.md +++ b/roles/openshift_metrics/README.md @@ -5,37 +5,81 @@ OpenShift Metrics Installation Requirements ------------ -It requires subdomain fqdn to be set. -If persistence is enabled, then it also requires NFS. + +The following variables need to be set and will be validated: + +- `openshift_metrics_hostname`: hostname used on the hawkular metrics route. + +- `openshift_metrics_project`: project (i.e. namespace) where the components will be + deployed. + Role Variables -------------- -From this role: +For default values, see [`defaults/main.yaml`](defaults/main.yaml). + +- `openshift_metrics_image_prefix`: Specify prefix for metrics components; e.g for + "openshift/origin-metrics-deployer:v1.1", set prefix "openshift/origin-". + +- `openshift_metrics_image_version`: Specify version for metrics components; e.g. for + "openshift/origin-metrics-deployer:v1.1", set version "v1.1". + +- `openshift_metrics_hawkular_cert:` The certificate used for re-encrypting the route + to Hawkular metrics. The certificate must contain the hostname used by the route. + The default router certificate will be used if unspecified + +- `openshift_metrics_hawkular_key:` The key used with the Hawkular certificate + +- `openshift_metrics_hawkular_ca:` An optional certificate used to sign the Hawkular certificate. + +- `openshift_metrics_hawkular_replicas:` The number of replicas for Hawkular metrics. -| Name | Default value | | -|-------------------------------------------------|-----------------------|-------------------------------------------------------------| -| openshift_hosted_metrics_deploy | `False` | If metrics should be deployed | -| openshift_hosted_metrics_public_url | null | Hawkular metrics public url | -| openshift_hosted_metrics_storage_nfs_directory | `/exports` | Root export directory. | -| openshift_hosted_metrics_storage_volume_name | `metrics` | Metrics volume within openshift_hosted_metrics_volume_dir | -| openshift_hosted_metrics_storage_volume_size | `10Gi` | Metrics volume size | -| openshift_hosted_metrics_storage_nfs_options | `*(rw,root_squash)` | NFS options for configured exports. | -| openshift_hosted_metrics_duration | `7` | Metrics query duration | -| openshift_hosted_metrics_resolution | `10s` | Metrics resolution | +- `openshift_metrics_cassandra_replicas`: The number of Cassandra nodes to deploy for the + initial cluster. +- `openshift_metrics_cassandra_storage_type`: Use `emptydir` for ephemeral storage (for + testing), `pv` to use persistent volumes (which need to be created before the + installation) or `dynamic` for dynamic persistent volumes. -From openshift_common: +- `openshift_metrics_cassandra_pv_prefix`: The name of persistent volume claims created + for cassandra will be this with a serial number appended to the end, starting + from 1. -| Name | Default Value | | -|---------------------------------------|----------------|----------------------------------------| -| openshift_master_default_subdomain | null | Subdomain FQDN (Mandatory) | +- `openshift_metrics_cassandra_pv_size`: The persistent volume size for each of the + Cassandra nodes. +- `openshift_metrics_heapster_standalone`: Deploy only heapster, without the Hawkular Metrics and + Cassandra components. + +- `openshift_metrics_heapster_allowed_users`: A comma-separated list of CN to accept. By + default, this is set to allow the OpenShift service proxy to connect. If you + override this, make sure to add `system:master-proxy` to the list in order to + allow horizontal pod autoscaling to function properly. + +- `openshift_metrics_startup_timeout`: How long in seconds we should wait until + Hawkular Metrics and Heapster starts up before attempting a restart. + +- `openshift_metrics_duration`: How many days metrics should be stored for. + +- `openshift_metrics_resolution`: How often metrics should be gathered. + +## Additional variables to control resource limits +Each metrics component (hawkular, cassandra, heapster) can specify a cpu and memory limits and requests by setting +the corresponding role variable: +``` +openshift_metrics_<COMPONENT>_(limits|requests)_(memory|cpu): <VALUE> +``` +e.g +``` +openshift_metrics_cassandra_limits_memory: 1G +openshift_metrics_hawkular_requests_cpu: 100 +``` Dependencies ------------ openshift_facts -openshift_examples + Example Playbook ---------------- |