サポートと今すぐチャット
サポートとのチャット

Foglight for Container Management 4.0 - User Guide

Understanding Foglight for Container Management Using Foglight for Container Management Reference

Load Balancer

Figure 22. Kubernetes Load Balancer Dashboard

The Load Balancer dashboard includes information about Kubernetes service, endpoint, and ingress. A Kubernetes ingress can provide load balancing, SSL termination, and name-based virtual hosting. A Kubernetes service is an abstraction which defines a logical set of pods and a policy by which to access them - sometime called micro-services. Kubernetes will update the endpoint whenever the set of pods in a service changes.

Alarms

Figure 23. Kubernetes Alarms Dashboard

The Alarms dashboard displays a list of alarms generated against the monitored Kubernetes environment. Use this view to quickly identify any potential problems related to a specific Kubernetes component.

Capacity Management

Foglight for Container Management provide capacity management feature for Kubernetes. This feature uses historical data to predict the trend and usage within a specific future period.

* 
NOTE: If the Capacity Management tab is not displayed, ensure the following:
1. You have purchased a license for Capacity Management. If not, contact Quest Support to purchase a license.
2. You have the Container Administrator role.
Figure 24. Capacity Management for Kubernetes

The Capacity Management dashboard contains the following fields:

•
Setting: Click to change the following values:
•
Baseline for Forecasting: Defines the historical period used for the calculations of metric views, current capacity, and recommended resources in the Resource Utilization view. The default value is 60 Days Trending.
•
Time Frame: Defines the predicted period for calculating metric views, current capacity, and recommended resources in the Resource Utilization view. The default value is Next 30 Days.
•
Resource Utilization:
•
Current Capacity: current resource capacity.
•
Current Performance: current resource usage.
•
Recommendation:
In this section, it shows how many resources are recommended to be added in the current trend, so as to meet the predicated usage.
•
Projected CPU/Memory Usage: Shows the historical data and the predicted usage trend within the configured future period.
•
Projected CPU/Memory Request: Shows the historical data and the predicted request trend within the configured future period.
•
Utilization: usage.
•
Capacity: upper bound which the usage might reach.
•
Growth Rate per Week: growth amount of the resource.
•
Time to Full: how many days the resource usage/request will reach the capacity.
* 
NOTE: If a value Never is displayed at Time to Full, which means the usage/request trend is declining and the usage/request will never reach the capacity.

Cost

Foglight for Container Management provide cost management feature for Kubernetes. The Cost view appears after clicking Container > Kubernetes > Cost. Currently, the Kubernetes costs for AKS and EKS environment are calculated.

* 
NOTE: If the Cost tab is not displayed, ensure the following:
1. You have purchased a license for Container Cost. If not, contact Quest Support to purchase a license.
2. You have the Container Administrator role.
Prerequisites

Container cost does not collect costs from cloud providers, but uses the cost data collected by cloud agents for parsing the Container environment cost. The cost start date is the first date of the current month.

To use the cost data, ensure the following:

•
For AKS cost, an Azure agent with the version 1.9.0 or above is required. The agent should be enabled to collect the inventory data of the AKS cluster resource group and that of the related infrastructure resource group, as well as the subscription cost related to the AKS cluster.
•
For EKS cost, an AWS agent with the version 1.9.0 or above is required. The agent should be enabled to collect the regional inventory data related to the EKS cluster and the account cost related to the EKS cluster.
* 
NOTE: Cost metrics should be configured for cloud agents to collect cost data. Refer to the Cost Tab section in Foglight Hybrid Cloud Manager User and Administration Guide for detailed information.
•
The Kubernetes agent collects the inventory and metrics data. Otherwise, the cost data won’t be displayed if there is no Kubernetes data on that day.
•
The clusters have been assigned to BUs. Otherwise, the clusters costs won’t be displayed in Cost Dashboard.
Administration dashboard

Administration dashboard enables you to manage the relationships between your cluster and BU, configure budgets at cluster level, and view summarized information about cluster, BU, and cost.

You need to assign cluster to BU here in order to see cost metrics aggregated at different levels in Container Cost dashboards.

Business Units tab

Business Units tab enables you to manage your BU for your Organizations. You can create different BUs to manage your clusters.

Figure 25. Business Units tab

The Business Units tab displays the following information:

•
Name: BU name.
•
Location: Geographic location of the BU.
•
Organization: Organization of this BU.
•
Clusters: numbers of the Kubernetes clusters that have been assigned to this BU.
•
Add: Add a new Business Unit and assign it to a specific organization.
•
Delete: Delete a Business Unit.
•
: Click the button to edit the information of this Business Unit.
•
Assign Organization: click the button to assign Business Unit to another Organization.
Clusters tab

Kubernetes clusters are collected by Kubernetes agents. By default, clusters will not be assigned to any BU and do not have a default monthly budget.

* 
NOTE: For EKS cluster, Container cost will automatically detect the EKS cluster collected by AWS agent for the cluster. However, if different AWS agent collects the same EKS cluster resources, then you need to assign the AWS account for this cluster.
Figure 26. Clusters tab

The Clusters tab displays the following information:

•
Name: The name of the Kubernetes cluster.
•
Business Unit: The name of the BU.
•
Platform: AKS, EKS, or On-Prem. If Container cost failed to detect the AKS or ESK cluster, it will display On-Prem by default.
•
Spending: The trend of cost spending.
•
Monthly Budget: Monthly budget, which can be configured through Set Budget.
•
Last Month (Month.Year): The cost of last month.
•
Month-to-Date (Month.Year): The cost of current month.
•
Forecast (Month.Year): The cost prediction for next month.
•
Last Collected Date: The latest date till when the cost data is collected. It depends on the Last collected time from different platforms.
When mousing over the time, the start date and the end date of container cost collection are displayed.
•
Set Budget: Set monthly budget for the cluster
•
Assign Business Unit: Assign the cluster to a specific BU.
•
Remove from Business Unit: Remove the cluster from the selected BU.
•
Assign AWS Account: This button will be enabled for editing only when an EKS cluster is selected.
•
: Click the button to change the monthly budge for the cluster.
Kubernetes dashboard

The Kubernetes dashboard displays a Cost Overview and several cost breakdown charts. The cost breakdown charts vary when choosing All Organizations, a single organization, a single BU, or a single cluster.

Kubernetes Overview
Figure 27. Kubernetes Overview

The Kubernetes overview shows the following information:

•
Time Range: Enables you to change the time range for displaying the cost, including This Month, Last Month, or Last N months.
•
Cost Overview: Monthly cost aggregated from all the clusters and the forecast aggregated cost for the continuing months.
•
Cost Breakdown By Organizations: Cost categories by different organizations. Displayed when choosing All Organization.
•
Cost Breakdown By Business Units: Cost categories by different BUs cross Organizations. Displayed when choosing All Organization or an organization.
•
Cost Breakdown by Platforms: Cost categories by different platforms, such as AKS, EKS, and so on.
•
Most Expensive Clusters: Displays the top N clusters that costs the most. Displayed when choosing an organization.
•
Clusters Cost: Cost of the clusters of this BU. Displayed when choosing a BU.
•
Alarms: all the cluster cost alarms. Currently we have over-budget alarm.
Cluster Overview dashboard
Figure 28. Cluster Overview

The Cluster overview dashboard shows the following information:

•
Time Range: Enables you to change the time range for displaying the cost, including This Month, Last Month, or Last N months.
•
Cluster Overview: The monthly cost aggregated from all the clusters of the organization and the forecast aggregated cost for the continuing months.
•
Top 5 Most Expensive Namespaces: Displays the top 5 most expensive namespaces whose cost is aggregated from its workloads.
•
Top 5 Most Expensive Workloads: Displays the top 5 most expensive workloads whose cost is aggregated from its related pods.
•
Top 5 Most Expensive Services: Displays the top 5 most expensive services whose cost is aggregated from its related pods.
•
Top 5 Most Expensive Nodes: Displays the top 5 most expensive nodes whose cost is collected from Azure or AWS.
•
Top 5 Most Expensive PVs: Displays the top 5 most expensive persistent volumes whose cost is collected from Azure or AWS.
•
Allocation Breakdown: Displays the Allocation Breakdown cost by Deployment or Stateful Set.
•
Alarms: Displays all the cluster cost over-budget alarms.
•
Cost Table button: Clicking the button will navigate to a Cost table grid layout dashboard.
•
View Topology button: Clicking the button will navigate to a Cluster Topology View.
Cost Table

The Cost Table lists the detailed cost information calculated by different categories, including Infrastructure, Workload, and Service.

Figure 29. Cost Table

•
Infrastructure: Lists the Azure or AWS costs collected.
•
Workload: The calculated workload cost is aggregated from its related pods. The workload types include deployment, statefulset, daemonset, and so on.
•
Service: The service cost includes related workload and load balancer costs in AWS.
View Topology

Click View Topology button to open the Cluster Topology View.

Figure 30. Cluster Topology View

Clicking virtual machine, a popup view will be displayed to show the pods cost which run on this node and also the persistent volume that used by some pod.

関連ドキュメント

The document was helpful.

評価を選択

I easily found the information I needed.

評価を選択