Creating a Dashboard

Creating a Dashboard

1. Log in to the AOM 2.0 console.
2. In the navigation pane, choose Dashboard.
3. Click next to Dashboard to create a dashboard group.
4. Click Add Dashboard in the upper left corner of the list.
5. In the displayed dialog box, set parameters.

   Table 1 Parameters for creating a dashboard

   Parameter	Description
   Dashboard Name	Name of a dashboard. Enter a maximum of 255 characters. The following special characters are not allowed: "$# %&'+;<=>?\
   Enterprise Project	Enterprise project. If you have selected All for Enterprise Project on the global settings page, select one from the drop-down list here. If you have already selected an enterprise project on the global settings page, this option will be dimmed and cannot be changed.
   Bind to Application	Select an application created in CMDB to bind. This configuration item is available only when the Application Insights function is enabled. To enable this function, see Configuring AOM Menus.
   Group Type	Options: Existing and New. Existing: Select an existing dashboard group from the drop-down list. New: Enter a dashboard group name to create one. Enter a maximum of 255 characters. The following special characters are not allowed: "$# %&'+;<=>?\

6. Click OK.

Adding a Graph to a Dashboard

After a dashboard is created, you can add graphs to the dashboard:

1. In the dashboard list, locate the target dashboard.
2. Go to the dashboard page, and select the Prometheus instance for which you want to add a graph from the drop-down list.
3. Go to the dashboard page. Click Add Graph or in the upper right corner to add a graph to the dashboard. For details about the graphs that can be added to the dashboard, see Graph Description. The data can be metric/log data. Select a graph as required.

   Table 2 Parameters for adding a graph

   Data Source	How to Add	Scenario
   Metric Sources	See Add a metric graph.	Monitors metrics of the business layer, application layer, Prometheus middleware, Prometheus running environments, Prometheus cloud services, open-source monitoring systems, Prometheus APIs/SDKs, and custom Prometheus plug-ins.
   Log Sources	See Add a log graph.	Monitors business metrics or other log metrics, such as latency, throughput, and errors cleaned based on ELB logs.

   • Add a metric graph. Set parameters by referring to Table 3. Then click Save.
     Figure 1 Adding a metric graph
     

     Table 3 Adding a metric graph

     Parameter	Description
     Graph Name	Name of a graph to distinguish it from other graphs. For graph names, variables can be added to dynamically filter graph information. Duplicate names are supported. Enter a maximum of 255 characters. The following special characters are not allowed: "$# %&'+;<=>?\
     Data Source	The default value is Metric Sources.
     Graph Type	Options: line, digit, top N, table, bar, and digital line.
     How to Add	Add metrics as required. You can select metrics from All metrics or using Prometheus statements.
     All metrics	Select target metrics from the metric drop-down list. Calculation method: Multiple Metrics: Performs calculation for metrics and their conditions separately, and displays the results on the graph. Combined Operations: Performs calculation on multiple metrics and their conditions based on expressions, and displays the results on the graph. Metric: Select a target metric from the drop-down list. You can also directly enter a metric name in the search box and click Generate. If no metric is reported, configure one. Statistical Period: Interval at which metric data is collected. The statistical periods that are available for you to select vary according to the time range. Relationship Between the Time Range and Statistical Period Condition: Metric monitoring scope. Each metric condition is in "key:value" format and can be selected from the drop-down list. You can also enter a dimension name and value, and click Generate to add a metric condition. You can also click and select AND or OR to add more conditions for the metric. Group Condition: Aggregate metric data by the specified field and calculate the aggregation result. Options: Not grouped, avg by, max by, min by, and sum by. For example, avg by clusterName indicates that metrics are grouped by cluster name, and the average value of the grouped metrics is calculated and displayed in the graph. Formatted Legend Name: Use a fixed name or variable as the legend name. Format: {{dimension name}}. If the displayed legend name is {{dimension name}}, there is no dimension. For example, enter {{hostname}} and a host name will be displayed as the legend name. Tables and digit/line graphs do not support Formatted Legend Name. You can click Add Metric to add more metrics. A maximum of 100 can be added.
     Prometheus statement	Add metric data by entering a Prometheus statement related to the metric. Prometheus Statement: See Prometheus Statements. Formatted Legend Name: Use a fixed name or variable as the legend name. If the displayed legend name is {{dimension name}}, there is no dimension. Format: {{dimension name}}. For example, enter {{hostname}} and a host name will be displayed as the legend name. To add more metrics, click Add Prometheus Statement. You can add 100 metrics in total.
     Graph Settings	On the right of the page, click the down arrow, select a desired graph type from the drop-down list, and set graph parameters (such as the X axis title, Y axis title, and displayed value). For details about the parameters, see Metric Data Graphs.
     Statistic	Method used to measure metrics. Options: Avg, Min, Max, Sum, and Samples.
     Time Range	Time range in which metric data is collected. Options: Last 30 minutes, Last hour, Last 6 hours, Last day, Last week, and Custom.
     Refresh Frequency	Interval at which the metric data is refreshed. Options: Refresh manually, 30 seconds auto refresh, 1 minute auto refresh, and 5 minutes auto refresh.

   • Add a log graph. Set parameters by referring to Table 4. Then click Save.
     Figure 2 Adding a log graph
     

     Table 4 Log graph parameters

     Parameter	Description
     Graph Name	Name of a graph to distinguish it from other graphs. Enter a maximum of 255 characters. The following special characters are not allowed: "$# %&'+;<=>?\
     Data Source	Click Log Sources.
     Log Group	Select a proper log group from the drop-down list box. If there is no log group you want to select, click Add Log Group to create one. For details, see Table 6.
     Log Stream	Select a proper log stream from the drop-down list. If there is no log stream you want to select, click Add Log Stream to create one. For details, see Table 6.
     Graph Settings	Select the required field from the structured field list and click next to the field name. Use the default SQL statements in the log graph or enter related query statements in the SQL statement query area as required. Specify the statistical period of log data. Options: Last minute, Last 5 minutes, Last 15 minutes, Last hour, Last 6 hours, Last day, Last week, or Custom. Click Execute Query to query related logs. By default, log data is displayed based on the graph type you set. You can select a graph type as required. Click to display the current log data in a table. Click to display the current log data in a bar graph. Click to display the current log data in a line graph. Click to display the current log data in a pie graph. Click to display the current log data in a number graph. Click to display the current log data in a digital line graph. Click to display the current log data in a national or provincial map. You can set the display parameters under a graph. For details, see Log Data Graphs.

4. Click . The graph is successfully added to the dashboard.

More Operations

After a dashboard is created, you can also perform the operations listed in Table 5.

Relationship Between the Time Range and Statistical Period

In AOM, a maximum of 1,440 data points can be returned for a single metric query. The relationship between the time range and statistical period is as follows:

Maximum time range = Statistical period x 1,440

If you select a time range shorter than or equal to the maximum time range, all the statistical periods that meet the preceding formula can be selected. For example, if you want to query metrics in the last hour, the available statistical periods are 1 minute, and 5 minutes.

For a dashboard, the relationship between the time range and statistical period is shown in the following table.