Dashboard parameters are input fields, such as drop-downs or text entries, you can use to narrow or customize the results displayed in the cards on your dashboard. For example, a dashboard parameter can be a text entry for an image hash, or a drop-down to select an instance name. These parameters can be referenced in OPAL scripts using a leading $, for example, $param.

Parameters can be of any basic data type, such as string, int64, float64, or bool. If you need to pass a non-storable value like a regular expression, you pass it as a string and cast it to the required type in your OPAL script.

Example usage in OPAL:

// Filter rows where the parameter $param is 'first' or 'second'
filter in($param, make_array('first', 'second'))

Parameters make dashboards interactive and flexible, allowing you to filter and explore data dynamically. In this example, there is a dashboard parameter configured and selected so that only data form the ai-api container is shown:

Types of dashboard parameters

The type of parameter you choose depends on what you want to do with the parameter.

Dashboard parameters that let you filter and pivot

If you want to add a parameter so that you can filter the data in your dashboard cards, and also pivot to other areas of Observe while carrying over those values to the destination for your pivot, choose one of the following:

Dashboard parameters that let you filter only

If you want to add a parameter just to filter the data in the cards on your dashboard, choose one of the following:

Create dashboard parameters

The flow for creating dashboard parameters varies depending on the parameter type:

• Single-value and multi-value parameters are not limited to a specific type of data. You can choose from Datasets, cards, or custom key/value pair lists as the source for the drop-down values.
• All other parameter types are limited to specific data type. Text and numbers are user-provided values, and correlation tags, resources, and metric tags get their values form their respective sources, so you don't get to choose the source for the parameter values.

The following examples of dashboard parameter creation are available:

• Create a multi-value dashboard parameter
• Create a correlation tag dashboard parameter
• Create a filtered Dataset dashboard parameter

Create a multi-value dashboard parameter

In this example, we will add a multi-value dashboard parameter that lets you filter the content in a single card. Using the My Kubernetes dashboard:

Perform the following steps to create a multi-value dashboard parameter.

1. Edit your dashboard, as shown in the example.

2. Click + Parameter.

3. Click Multi-value.

4. Give the parameter a name you want displayed on the dashboard, such as Container. By default, the OPAL token $Container is used, unless you want to change it. From the list of sources, select Dataset, then find and select the Kubernetes Logs Dataset. Lastly, select the container column.

   

5. Click Save & select cards.

6. Select the card and field to which you want to bind your parameter field. In this case

7. For each card, select the the card field you want to bind to your parameter field. In this case, we will bind the container parameter field to the container field in the card. The corresponding OPAL behind the scenes is container = $Container.

   

8. Click Save.

The dashboard now includes a dashboard parameter filter. Click on the filter, then find a select a container name, such as controller. The dashboard card is now updated so that only data from the controller container is shown.

You can select multiple values in a multi-value dashboard parameter, as shown below:

Create a correlation tag dashboard parameter

Perform the following steps to create a correlation tag dashboard parameter. In this example, we have a dashboard showing a variety of Kubernetes data from logs, metrics, and traces, as correlation tag parameters enable you to filter across different data sources. Notice that there are several cards grouping the data by namespace:

We will create a dashboard parameter to bind the k8s.namespace.name correlation tag to the correlation tag of the same name for the four cards also grouping by namespace.

1. Edit your dashboard.

2. Click + Parameter.

3. Click Correlation tag.

4. Select the k8s.namespace.name correlation tag. In the Display name field, you can change the value to use a more friendly label to display in the UI. We will use Namespace (k8s.namespace.name) as the display name.

   

5. Click Save & select cards.

6. Select the cards you want to apply this dashboard parameter to. We will select the four cards with the word Namespace in the card title. After you select the cards, you can see the parameter is automatically bound to the matching tag name in the cards. If this automatic binding doesn't happen you can click in the drop-down list and select the card column you want to bind to the correlation tag.

   

7. Click Save.

The dashboard now has a dashboard parameter configured. Click in the Namespace (k8s.namespace.name) drop-down list and select a value to filter the cards. For example, if we select staging as the namespace, we see in those four cards, only data from staging is shown:

Correlation tag parameters make dashboards available as a destination for pivoting. For example, if we look at Kubernetes Logs from the Log Explorer, we can pivot from k8s.namespace.name to any dashboards also containing the same correlation tag dashboard parameter.

In the namespace column, click on the down caret icon (), then select Dashboard. In this example, we select the eng namespace to pivot from:

Next, you see a list of dashboards containing the k8s.namespace.name parameter, along with related parameters. We want to pivot to Ks K8s Dashboard, which we were working with earlier, so select that in the list of dashboards:

In the Ks K8s Dashboard, you can see the four cards where we have applied the dashboard parameter now only show data from the eng namespace, which is where we pivoted from in the Log Explorer.

We can now set this dashboard as the default landing page any time anyone pivots on our correlation tag. Click the vertical ellipsis icon (), then select Set as default, then select the resource or correlation tag. In our case, we select the k8s.namespace.name correlation tag.

📘

Note

Setting a default dashboard applies to the entire workspace and not just for the individual user.

Create a filtered Dataset dashboard parameter

Create a filtered Dataset dashboard parameter to filter the cards on your dashboard using a single Dataset. In our example, we can tie the Kubernetes Logs Dataset to the Ks K8s Dashboard, and use the fields from the Dataset such as container, cluster, or pod to filter the cards on our Dashboard.

Perform the following steps:

1. Edit your dashboard.
2. Click + Parameter.
3. Click Filtered Dataset.
4. Click the Select source drop-down and select Kubernetes Logs as the Dataset. Leave the default Apply filtering to all relevant cards option selected, to let the Observe platform determine which cards are relevant and should have this filter applied.
5. Click Apply.

You can see a new filter dashboard is available now, populated with the fields from the Dataset, such as stream, cluster, and namespace.

To use the filter, select any options in the filter dashboard to apply them as filters to the cards in your dashboard. For example, we select the prod namespace, and see that the Log Volume by Namespace Over Time card is filtered to show only data from prod:

In addition to using the filter dashboard, the filtered Dataset also provides the full filter bar experience, where you can enter your own text and operators, or click in the filter bar and use the help to assist you in building your queries.