Skip to main contentCloud Pak Deployer Monitor

Cloud Pak for Data Watson Machine Learning Development Space Information

Monitor Summary

This monitor will generate the following observations:

  • Number of Cloud Pak for Data Waston Machine Learning Deployment Spaces
  • Number of Cloud Pak for Data Waston Machine Learning Deployments per space
  • Number of Failed Cloud Pak for Data Waston Machine Learning Deployments per space
  • Number of Online Cloud Pak for Data Waston Machine Learning Deployments per space
  • Number of Batch Cloud Pak for Data Waston Machine Learning Deployments per space

Monitor Requirements

In order for Watson Studio Job Information to be available, CCS (Common Core Services) must be available on the Cloud Pak for Data instance. Catridges like Watson Studio and Watson Machine Learning automatically install CCS as part of their dependencies. If CCS is not available, no watson studio job information is available.

Administrative Credential Requirements

In order to retrieve the monitor data, the credential of admin user will be used. The default password of admin user is saved in the secret admin-user-details. You could use the below command to query the default password of the admin user..

oc extract secret/admin-user-details --keys=initial_admin_password --to=-

Once the password of the admin user is changed, please also change the password in the secret admin-user-details.

oc set data secret/admin-user-details initial_admin_password=<new password>

Deployment of the monitor

This monitor can be deployed using the IBM Cloud Pak Deployer or manually deployed using oc commands:

Monitor Source

Source folder containing the script can be found here (cp4d-wml-deployment-space-info).

Generated observations

Once the monitor is deployed, the following observations are available in IBM Cloud Pak for Data Metrics:

Using the IBM Cloud Pak for Data Platform Management Events:

https://<CP4D-BASE-URL>/zen/#/platformManagement/events

On the Platform Management Events page the following entries are added:

EventEvent TypePossible valuesDescription
Cloud Pak for Data Waston Machine Learning Deployment Space Info Overall CountNumber of Cloud Pak for Data Waston Machine Learning Deployment SpacesinfoThe Cloud Pak for Data Waston Machine Learning Spaces information are requested.
Cloud Pak for Data Waston Machine Learning Deployment Space Deployed CountNumber of Cloud Pak for Data Waston Machine Learning Deployments per spaceinfoThe Cloud Pak for Data Waston Machine Learning Spaces information are requested.
Cloud Pak for Data Waston Machine Learning Deployment Space Deployed Falied CountNumber of Failed Cloud Pak for Data Waston Machine Learning Deployments per spaceinfoThe Cloud Pak for Data Waston Machine Learning Spaces information are requested.
Cloud Pak for Data Waston Machine Learning Deployment Space Deployed Count OnlineNumber of Online Cloud Pak for Data Waston Machine Learning Deployments per spaceinfoThe Cloud Pak for Data Waston Machine Learning Spaces information are requested.
Cloud Pak for Data Waston Machine Learning Deployment Space Deployed Count BatchNumber of Batch Cloud Pak for Data Waston Machine Learning Deployments per spaceinfoThe Cloud Pak for Data Waston Machine Learning Spaces information are requested.
Overview Events and Alerts

Using the IBM Cloud Pak for Data Prometheus endpoint

https://<CP4D-BASE-URL>/zen/metrics

It will generate 2 types of metrics:

  • Actual Result metrics
    These metrics contain the observed values
  • Zen Watchdog metrics
    These metrics are used by the Cloud Pak for Data Zen Watchdog to monitor the state and trigger notifications

Actual Result metrics:

  • cp4dplatformwmldeploymentspaceinfo_overall_count Provides number of Cloud Pak for Data Waston Machine Learning Deployment Spaces
  • cp4d_wml_deployment_space_deployed_count Provides number of Cloud Pak for Data Waston Machine Learning Deployments per space
  • cp4d_wml_deployment_space_deployed_fail_count Provides number of Failed Cloud Pak for Data Waston Machine Learning Deployments per space
  • cp4d_wml_deployment_space_deployed_count_online Provides number of Online Cloud Pak for Data Waston Machine Learning Deployments per space
  • cp4d_wml_deployment_space_deployed_count_batch Provides number of Batch Cloud Pak for Data Waston Machine Learning Deployments per space
# HELP cp4d_wml_deployment_space_deployed_count
# TYPE cp4d_wml_deployment_space_deployed_count gauge
cp4d_wml_deployment_space_deployed_count{event_type="cp4d_wml_deployment_space_deployed_count",monitor_type="cp4dplatformwmldeploymentspaceinfo",reference="test-demo"} 2
# HELP cp4d_wml_deployment_space_deployed_count_batch
# TYPE cp4d_wml_deployment_space_deployed_count_batch gauge
cp4d_wml_deployment_space_deployed_count_batch{event_type="cp4d_wml_deployment_space_deployed_count_batch",monitor_type="cp4dplatformwmldeploymentspaceinfo",reference="test-demo"} 1
# HELP cp4d_wml_deployment_space_deployed_count_online
# TYPE cp4d_wml_deployment_space_deployed_count_online gauge
cp4d_wml_deployment_space_deployed_count_online{event_type="cp4d_wml_deployment_space_deployed_count_online",monitor_type="cp4dplatformwmldeploymentspaceinfo",reference="test-demo"} 1

Zen Watchdog metrics (used in platform management events)

  • watchdog_cp4dplatformwmldeploymentspaceinfo_cp4dplatformwmldeploymentspaceinfo_overall_count
  • watchdog_cp4dplatformwmldeploymentspaceinfo_cp4d_wml_deployment_space_deployed_count
  • watchdog_cp4dplatformwmldeploymentspaceinfo_cp4d_wml_deployment_space_deployed_fail_count
  • watchdog_cp4dplatformwmldeploymentspaceinfo_cp4d_wml_deployment_space_deployed_count_online
  • watchdog_cp4dplatformwmldeploymentspaceinfo_cp4d_wml_deployment_space_deployed_count_batch

Zen Watchdog metrics can have the following values:

  • 2 (info)
  • 1 (warning)
  • 0 (critical)
# HELP watchdog_cp4dplatformwmldeploymentspaceinfo_cp4d_wml_deployment_space_deployed_count
# TYPE watchdog_cp4dplatformwmldeploymentspaceinfo_cp4d_wml_deployment_space_deployed_count gauge
watchdog_cp4dplatformwmldeploymentspaceinfo_cp4d_wml_deployment_space_deployed_count{event_type="cp4d_wml_deployment_space_deployed_count",monitor_type="cp4dplatformwmldeploymentspaceinfo",reference="test-demo"} 2
# HELP watchdog_cp4dplatformwmldeploymentspaceinfo_cp4d_wml_deployment_space_deployed_count_batch
# TYPE watchdog_cp4dplatformwmldeploymentspaceinfo_cp4d_wml_deployment_space_deployed_count_batch gauge
watchdog_cp4dplatformwmldeploymentspaceinfo_cp4d_wml_deployment_space_deployed_count_batch{event_type="cp4d_wml_deployment_space_deployed_count_batch",monitor_type="cp4dplatformwmldeploymentspaceinfo",reference="test-demo"} 2
# HELP watchdog_cp4dplatformwmldeploymentspaceinfo_cp4d_wml_deployment_space_deployed_count_online
# TYPE watchdog_cp4dplatformwmldeploymentspaceinfo_cp4d_wml_deployment_space_deployed_count_online gauge
watchdog_cp4dplatformwmldeploymentspaceinfo_cp4d_wml_deployment_space_deployed_count_online{event_type="cp4d_wml_deployment_space_deployed_count_online",monitor_type="cp4dplatformwmldeploymentspaceinfo",reference="test-demo"} 2