Logging is here!

Dashboards

To create a Timeboard or a Screenboard Select which of these you would prefer to create after clicking ‘New Dashboard’ in the ‘Dashboards’ dropdown.

Dashboard selection

What is the difference between a Screenboard and a Timeboard?

At Datadog we give you the capability to create and customize two types of dashboards; Screenboards and Timeboards. To more clearly understand the differences between the two, consider the following:

Timeboards Screenboards
Time Scope All graphs share same time scope All graphs can have individual time scope
Layout Graphs appear in a fixed grid Graphs are placed anywhere you like on the canvas
Can Share Graphs Individually Yes No
Can Share the Entire Dashboard No Yes
Sharing can be Read-Only Yes Yes

Editing template variables

Dashboard templating allows you to create dashboards that use variables like $scope or $redis in place of specific tags or hosts. You can then dynamically explore the metrics across different sets of tags. Select a new variable value in the dropdown menu, and that value applies across the dashboard.

To create, edit, and delete template variables click the gear icon at the upper right-hand side of the screen, then select ‘Edit Template Variables’ from the actions menu.

edit template variable

This open the template variable editing panel.

Redis template var

A template variable is defined by a name and optional parameters for ‘Tag Group’ and ‘Default Tag.’ A tag group is a prefix shared among several tags, like redis_port for the tags redis_port:6379 and redis_port:6380. Setting a tag group eliminates irrelevant tags from the variable’s scope selector, and removes the prefix from the listed values for clarity - so you’ll see 6379 and 6380 in the ‘Default Tag’ dropdown instead. The ‘Default Tag’ option determines the initial value for the variable on dashboard load.

Using template variables in graph editors/widgets

Redis-tpl graph editor

Once defined, template variables appear alongside normal tag and host options in graph editors. If you set 6379 as the value of $redis, all graphs defined with $redis is scoped to redis_port:6379.

Redis tpl selected

You can also use them explicitly in widgets such as Event Stream, with a query of the form tags:$redis.

Event Correlation at Design Time

Event Correlation refers to overlaying events on top of a dashboard graph and is an important feature of the Datadog platform. You can setup correlation at two different times: either when you setup the dashboard or adhoc at the time you view the dashboard.

guides-eventcorrelation-screenboard

Setup event correlation at design time by editing any graph on both Timeboards and Screenboards and adding events to the graph. To learn more about this, visit the Graphing Primer. You can find details about adding events using the UI or via the JSON interface further down the page.

Event Correlation at View Time

guides event correlation

Setup event correlation at view time by adding a query in the Search box at the top left of any Timeboard dashboard window. This replaces any events added at design time, but applies the events to all graphs on that particular dashboard.

Correlation between Logs and Metrics

Jump from a metric to its logs

Fast and easy correlation is key when troubleshooting an issue. Use the following shortcut from any dashboard timeseries graphs to open a contextual menu with the most related logs.

Related logs

Select View related logs to jump to the log explorer page zoomed on the selected timeframe with all the current context of your graph.

How do we define the search query ?

To define the most related logs, we use the following parameters:

  • Timeframe: Focused on the selected data point and uses the graph bucket size to display data before and after the selected point.
  • Integration prefix: If the metric is coming from an integration, Datadog filters on the source attribute with the same integration name.
  • Tags: All tags used in the graph (template variable, split by, filter by) are automatically added to the search query.

Further Reading