> ## Documentation Index
> Fetch the complete documentation index at: https://docs.chronosphere.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Visualize dashboard contents with panels

*Panels* allow you to visualize, filter, and transform the results of a query.
[Standard dashboards](/observe/dashboards) provide visualizations designed to work
with other Chronosphere Observability Platform features, such as Metrics Explorer.

<Note>
  [Classic dashboards](/observe/dashboards/classic-dashboards) have their own panels
  and methods of configuring them. For details about panels in classic dashboards,
  see [Classic dashboard panels](/observe/dashboards/classic-dashboards/classic-panels).
</Note>

## Common panel elements

All panels consist of common settings and features. You can add these during
panel creation, or edit them at a later time.

* **Name**: The panel's configured name.
* **Group**: The panel group that contains the panel.
* **Description**: A description of the panel displayed when holding the pointer over
  an Information icon.
* **Type**: A panel's [*type*](#panel-types) defines its visualization method or
  content. By default, new panels are the
  [Time Series Chart](/observe/dashboards/panels/time-series-chart) type.
* **<Icon icon="link-2" alt="Link icon" /> Links**: An icon that reveals any links
  configured for the panel.
* **<Icon icon="maximize-2" alt="Expand arrow icon" /> Expand**: Expands the
  panel to a the full width and height of the browser for a more detailed view.
* **<Icon icon="ellipsis-vertical" alt="Options icon" /> Options**: Opens a menu
  of panel-related actions, depending on the panel type.
  * **Edit panel**: Used to update the panel.
  * **Open in Metrics Explorer**: Click to open a new tab using the panel's queries
    in [Metrics Explorer](/investigate/querying/metrics/explorer). When opening
    Metrics Explorer from this link, **Resolution** settings carry over.
  * **Copy as link**: Copy a link directly to this panel. When used, the link
    opens the panel in full-screen mode.
  * **Copy as image**: Copy the panel as an image to the operating system clipboard.
  * **Export as PNG**: Download the panel as a PNG file. The filename matches the
    panel name.
  * **Add to dashboard**: Copy this panel to another dashboard. Select an existing
    dashboard or create a new one.
  * **Add to notebook**: See [Notebooks](/navigate/notebooks#from-another-page).
  * **Analyze anomaly (DDx)**: On time series panels, use this link to open the
    [metrics differential diagnosis](/investigate/analyze/differential-diagnosis/metrics)
    page with the panel's query.
  * **Inspect panel**: Review statistics about the panel's query or queries.
* **Pin**: Some charts let you pin data or a point in time. On supported charts, a
  tooltip appears when you hold your pointer over the chart. Click a point in the
  chart to pin. See individual chart pages for pin and tooltip options.

Panels can also have additional options presented in tabs of the **Edit panel** interface.
Most panels provide these tabs:

* **Query**: Observability Platform uses results from
  [queries](#change-a-panels-query) to populate the panel's visualization.
* **Settings**: Configuration options for the selected panel type.
* **Links**: [Links](#add-links-to-a-panel) associated with the panel, which can
  optionally use variables provided by Observability Platform related to the panel or
  dashboard.
* **JSON**: An editable, automatically updated
  [JSON representation](/observe/dashboards/panels/json-representation) of the panel's
  configuration.

### Time ranges

Panels react to the time range set for a dashboard, and can be dynamically customized
by using [dashboard variables](/observe/dashboards/customization/dashboard-variables).

Use a dashboard's *time range* to analyze visualized data in multiple
[panels](/observe/dashboards/panels). You can define a refresh interval to control
how frequently the dashboard updates panel data. You can also select a
[time range](/navigate/time-ranges) to view the same period across all of a
dashboard's panels to help you correlate events. You can also compare current data
to past data to identify trends.

## Create panels in a dashboard

To add and use panels in a dashboard, you must first
[create a dashboard](/observe/dashboards#create-a-new-dashboard) or select an existing
dashboard to which you'll add the panel.

Each panel in a dashboard belongs to a
[*panel group*](/observe/dashboards/customization/layouts#add-a-panel-group-to-a-dashboard),
which is a customizable container of one or more panels.
If a dashboard doesn't have any panel groups, Observability Platform creates one
for you when you add a panel.

<Tabs>
  <Tab title="Web" id="add-panels-to-a-dashboard-web">
    To add a panel to a standard dashboard with no existing panel groups:

    1. Open the dashboard. You can
       [view a list of all dashboards](/observe/dashboards#view-dashboards) or select
       from a [list of dashboards in a collection](/administer/collections/home).
    2. Click **<Icon icon="pencil" alt="Edit icon" /> Edit** on the dashboard.
    3. Click **<Icon icon="chart-column" alt="Panel icon" /> Panel**.
    4. In the **Add Panel** interface that appears, provide a **Name** for the new panel.
    5. Optional: Configure the panel based on the [panel options](#common-panel-elements) and
       [panel type](#panel-types) you want to use.
    6. Click **Add** to add the panel or **Cancel** to discard it.

    Users with
    [Generative AI](/overview/generative-ai#generative-ai-features-in-observability-platform)
    features enabled can generate the panel **Name** and **Description** based on the
    panel contents. After creating the panel, edit the panel and click **Generate**.

    To add a new panel to a panel group:

    1. In the panel group to which you want to add a panel, click the
       **<Icon icon="chart-column" alt="Add panel to group icon" /> Add panel to group**
       icon.
    2. In the **Add Panel** interface that appears, provide a **Name** for the new panel.
    3. Click **Add** to add the panel or **Cancel** to discard it.

    Panels aren't permanently added to a dashboard until you save all changes to
    the dashboard by clicking **Save** on the dashboard.

    You can save a new panel to a dashboard with as little as a name, then edit
    it later to configure it and add or modify queries.
  </Tab>

  <Tab title="Chronoctl" id="add-panels-to-a-dashboard-chronoctl">
    Chronosphere Observability Platform imports, exports, and updates dashboards as JSON
    representations, which you can submit to Observability Platform as an inline string value in
    the [Chronoctl](/tooling/chronoctl) YAML representation of a dashboard resource.

    For details about creating and updating dashboard JSON in Chronoctl YAML, see
    [Create a dashboard](/observe/dashboards#create-a-new-dashboard). For details about the panel
    JSON representation format, see
    [JSON representation](/observe/dashboards/panels/json-representation).
  </Tab>
</Tabs>

### Panel types

For details about these settings or unique features in a panel type, see its
documentation. The panel types available for visualizing data include:

* [**Time Series Chart**](/observe/dashboards/panels/time-series-chart): A line
  or bar chart representation of time series data returned by the panel's queries.
* [**Bar Chart**](/observe/dashboards/panels/bar-chart): A bar chart representation
  of data returned by the panel's queries.
* [**Heat Map**](/observe/dashboards/panels/heat-map): The density of data points
  across ranges of values returned by a query, visualized by colors that represent
  the magnitude of data for each range of values.
* [**Status History**](/observe/dashboards/panels/status-history): Status over time
  as rows of boxes, with colors from numeric thresholds or label-based color
  mappings.
* [**Stat Chart**](/observe/dashboards/panels/stat-chart): A stat chart visualizes
  statistic numbers returned by a query.
* [**Gauge Chart**](/observe/dashboards/panels/gauge-chart): A value for each data
  point returned by a query, visualized as a configurable gauge depicting a minimum
  and maximum range and configurable thresholds.
* [**Markdown**](/observe/dashboards/panels/markdown): Renders defined Markdown
  text.
* [**External**](/observe/dashboards/panels/external): Embed external resources
  using an iframe.
* [**Table**](/observe/dashboards/panels/markdown): Presents the data returned by
  the query as a customizable table.
* [**Pie Chart**](/observe/dashboards/panels/pie-chart): A pie chart representation
  of data returned by the panel's queries.
* [**Distribution Chart**](/observe/dashboards/panels/distribution-chart): A
  visualization of histograms by bucket.
* [**Service Topology**](/observe/dashboards/panels/service-topology): View trace
  data related to this dashboard.

<Warning>
  All panel types except Markdown, Service topology, and External use queries.

  Queries that return an extremely large number of data points or invalid results
  can result in panel errors. For example, a query might return an error for exceeding
  server resource limits.

  Observability Platform reports these errors with an icon that appears in the corner
  of the **Preview** pane of the **Add panel** or **Edit panel** interfaces, or on the
  panel when viewing it on the dashboard. Hold the pointer over the icon to view the
  error message.
</Warning>

## View panel information

The **Inspect panel** drawer presents information about your panel. Panel inspection
isn't valid for Markdown or Service Topology panels.

### Overview

The **Overview** tab displays statistics about the performance of your query or queries.
Overview details vary depending on the [panel type](#panel-types). All panels have
the following options:

* **Number of queries**: How many queries the panel uses to generate the
  visualization.
* **Query type**: The type of query, which is related to the panel type.
* **Longest request duration**: The longest response time for a query request.

Time series charts have these additional statistics:

* **Interval step size**: The query [interval size](/observe/dashboards/customization/intervals).
* **Number of series**: The number of time series returned.
* **Max data points per series**: The maximum number of data points returned per series.
* **Total number of data points**: Number of data points returned by the query.

Table charts have these additional statistics:

* **Number of table rows**: For table charts, the number of rows in the table.
* **Number of table columns**: For table charts, the number of columns in the table.

Bar, stat, gauge, pie, and distribution charts have this additional statistic:

* **Number of series (single value)**: The number of series returned.

### Queries

The **Queries** tab displays details for each query used to generate the panel. Each
query has its own informational box, such as **Query 1** and **Query 2**. Each query
has the following entries:

* **Query**: The query you entered, including any
  [variables](/observe/dashboards/customization/dashboard-variables) used.
* **Raw Query**: The actual query used to retrieve data, which translates variables
  into real values.

## Edit a panel

You can customize the features of each panel in a dashboard and edit the queries
used to populate it.

<Tabs>
  <Tab title="Web" id="edit-a-panel-web">
    To edit a panel in a standard dashboard:

    1. Open the dashboard. You can
       [view a list of all dashboards](/observe/dashboards#view-dashboards) or select
       from a [list of dashboards in a collection](/administer/collections/home).
    2. Click **Edit** for the dashboard.
    3. Click **Edit** for the panel you want to edit.
    4. In the **Edit panel** interface that appears, configure the panel based on the
       [panel options](#common-panel-elements) and [panel type](#panel-types) you want to
       use. Users with
       [Generative AI](/overview/generative-ai#generative-ai-features-in-observability-platform)
       features enabled can click **Generate** to update the panel **Name** and
       **Description** based on the panel contents.
    5. Click **Apply** to apply the changes or **Cancel** to discard them.

    After you've finished editing panels, click **Save** on the dashboard to save your
    changes.

    To change the layout of the edit panel page, click a **Preview layout** button next
    to the time range selector.
  </Tab>

  <Tab title="Chronoctl" id="edit-a-panel-chronoctl">
    Chronosphere Observability Platform imports, exports, and updates dashboards as JSON
    representations, which you can submit to Observability Platform as an inline string value in
    the [Chronoctl](/tooling/chronoctl) YAML representation of a dashboard resource.

    For details about creating and updating dashboard JSON in Chronoctl YAML, see
    [Create a dashboard](/observe/dashboards#create-a-new-dashboard). For details about the panel
    JSON representation format, see
    [JSON representation](/observe/dashboards/panels/json-representation).
  </Tab>
</Tabs>

### Change a panel's type

You can change the type of panel at any time.

<Tabs>
  <Tab title="Web" id="change-a-panels-type-web">
    To change a panel's type:

    1. Open the panel's editor.
    2. Click the **Type** dropdown.
    3. Select a panel type.
    4. Click **Apply** to save any panel type changes and exit the panel editor.
  </Tab>

  <Tab title="Chronoctl" id="change-a-panels-type-chronoctl">
    Chronosphere Observability Platform imports, exports, and updates dashboards as JSON
    representations, which you can submit to Observability Platform as an inline string value in
    the [Chronoctl](/tooling/chronoctl) YAML representation of a dashboard resource.

    For details about creating and updating dashboard JSON in Chronoctl YAML, see
    [Create a dashboard](/observe/dashboards#create-a-new-dashboard). For details about the panel
    JSON representation format, see
    [JSON representation](/observe/dashboards/panels/json-representation).
  </Tab>
</Tabs>

### Change a panel's query

Queries populate panels with data to visualize. Use one of the following methods
to create a query:

* In the **Code** tab, write your own query or paste a known query.
* Use the **Builder** tab to create a query using the
  [query builder](/investigate/querying/metrics/query-builder).

<Tabs>
  <Tab title="Web" id="change-a-panels-query-web">
    You can change the query a panel uses when the data doesn't match your needs.

    1. Open the panel's editor.

    2. Select the **Query** tab.

    3. Optional:
       * Change the **Resolution** of a query. This setting controls the maximum
         data points per series. Options are:
         * **Low**: Prioritizes query performance over accuracy.
         * **Medium**: (Default) Balances performance and accuracy.
         * **High**: Uses more data points to provide more precise results, but might have
           degraded performance.
         * **Custom**: Set a custom step size. If you select **Custom**, you must enter a
           **Custom resolution** with [time units](/overview/concepts/time-units). For
           example, `30m`. This field supports duration
           [variables](/observe/dashboards/customization/dashboard-variables). For more
           details, see [Dashboard query interval and resolution](/observe/dashboards/customization/intervals).
         * Change the **Time preference**:
           * Toggle **Override dashboard time**.
           * Add a **Relative time override**. You can use custom variables in the
             override if they're duration-based.

    4. Select a **Datasource** for the query. This selection determines the syntax you
       can enter. For example, selecting **Metrics (Prometheus)** accepts
       [PromQL syntax](/investigate/querying/promql), whereas selecting **Logs** accepts
       [Chronosphere log syntax](/investigate/querying/query-logs/query-syntax).

    5. Enter expressions for the query you want to edit.

       The visualization preview changes as you edit the query. If the query doesn't
       return any data, the visualization preview doesn't indicate any changes. If the
       query or results return an error, an icon appears next to the panel name.

    6. Optional: Add a **Series alias** to define the name displayed for the series
       in the panel's tooltip and legend.

           <Note>
             If you select a **Logs** query that includes a field which contains a period (`.`)
             in its name (such as `k8s.namespace`), and you want to use signals to group
             notifications, use an
             [alias for that field name](/investigate/querying/query-logs/query-syntax#alias-field-names).
             Otherwise, periods are converted to underscores in the generated visualization.
           </Note>

    7. Optional: Change the **Resolution** of a query. This setting controls the maximum
       data points per series. Options are:
       * **Low**: Prioritizes query performance over accuracy.
       * **Medium**: (Default) Balances performance and accuracy.
       * **High**: Uses more data points to provide more precise results, but might have
         degraded performance.
       * **Custom**: Set a custom step size. If you select **Custom**, you must enter a
         **Custom resolution** with [time units](/overview/concepts/time-units). For
         example, `30m`. This field supports duration
         [variables](/observe/dashboards/customization/dashboard-variables).

         Lower resolutions will return results faster, but can reduce data spikes due to
         [downsampling](/control/shaping/shape-metrics/downsampling). If you find
         you're missing spikes at **Low** resolution, changing to **High** resolution
         or using an aggregation function like `max_over_time` can help. See
         [Disappearing spikes when increasing the range​](https://chronosphere.io/learn/understanding-promql-and-its-quirks/#Disappearing%20spikes%20when%20increasing%20the%20range%E2%80%8B)
         in the Chronosphere blog for a detailed explanation.

    8. Optional: Add additional queries by clicking **Add query**.

    9. Optional: Hide a visualized query by clicking **Hide query**, or show a hidden
       query by clicking **Show query**.

    10. Change a query's **Options** based on the [panel type](#panel-types).

    11. Click **Apply** to save any query changes and exit the panel editor.

    For details about writing queries, read the [querying documentation](/investigate/querying).

    You can use [dashboard variables](/observe/dashboards/customization/dashboard-variables)
    in a query to let dashboard users dynamically customize them without modifying
    the query itself. See
    [Apply a variable to multiple panels](/observe/dashboards/customization/dashboard-variables#apply-a-variable-to-multiple-panels)
    to add one variable to supported queries on many panels at once.

    #### Variable suggestions

    Variable suggestions appear in the panel query editor only when editing a panel query on
    a standard dashboard using the **Web** tab steps in this section. Variable suggestions
    don't appear in Metrics Explorer or other standalone query editors.

    Variable suggestions apply only when **Datasource** is **Metrics (Prometheus)** and the
    panel query is PromQL. Observability Platform doesn't show variable suggestions for other
    telemetry [data types](/overview/types).

    When PromQL in the panel includes `$<NAME>` and no dashboard variable is defined for that
    name, the panel query editor shows the following text:

    ```text wrap theme={null}
    Suggestion: The following variables are not configured in your dashboard. Would you like to configure and add them?
    ```

    Each undefined name appears as its own link, such as **`$service`**. Click a link to
    add that variable to the dashboard and configure it.

    When the panel query editor can replace a literal label with an existing dashboard
    variable, a row starts with **Suggested variables:**, followed by one or more
    matching labels in the form `label=~"$<NAME>"`. In that pattern, `label` is your
    Prometheus label and `<NAME>` is the dashboard variable name. Each matching label is
    a link, which you can click to insert that label text into the query.

    After each matching label, the panel query editor shows a short note in parentheses
    with the following details:

    * **seen in 1 other query** when exactly one other query on the same dashboard already
      uses that variable on that label.
    * **seen in N other queries** when more than one other query on the same dashboard already
      uses that variable on that label.
    * **label match** when the variable name matches the label in your literal matcher
      and Observability Platform doesn't show a count from other dashboard queries.
  </Tab>

  <Tab title="Chronoctl" id="variable-suggestions-chronoctl">
    Chronosphere Observability Platform imports, exports, and updates dashboards as JSON
    representations, which you can submit to Observability Platform as an inline string value in
    the [Chronoctl](/tooling/chronoctl) YAML representation of a dashboard resource.

    For details about creating and updating dashboard JSON in Chronoctl YAML, see
    [Create a dashboard](/observe/dashboards#create-a-new-dashboard). For details about the panel
    JSON representation format, see
    [JSON representation](/observe/dashboards/panels/json-representation).
  </Tab>
</Tabs>

### Add links to a panel

To add links to a panel:

1. Open the panel's editor.

2. Click **Links**.

3. Click **Add link**.

4. Choose a **Link type**, and then configure it:

   <Tabs>
     <Tab title="Dashboard" id="dashboard-link">
       If you select **Dashboard**:

       1. Click the **Select dashboard** box and choose a dashboard.
       2. Select or clear the following options:
          * **Include current time range**
          * **Include current variable values**
     </Tab>

     <Tab title="Link" id="link-link">
       If you select **Link**:

       1. Enter a **Name** for the link, which the panel uses for the link's text.
       2. Enter a **URL** for the link's destination.

          You can include dashboard variables in the URL. Chronosphere includes available
          variables in the link insertion dialog.
     </Tab>
   </Tabs>

5. Click **Save link** to save the link.

### View a panel's JSON

To view the JSON for a single panel:

1. Open the panel's editor.
2. Select the **JSON** tab.

### Exit the panel editor

To save changes to a panel and exit the panel editor, click **Apply**.

To exit the panel editor without saving changes, click **Cancel**.

## Delete a panel

To delete a panel:

1. Open the dashboard you want to change.
2. In the panel you want to delete, click the <Icon icon="trash" />**Delete**
   icon.
3. Click **Delete** to confirm or **Cancel** to go back.

## Save changes to a dashboard

Changes made to panels in a dashboard aren't saved until you save all changes to
the dashboard, even after applying the changes to the panels in the **Edit panel** interface.

To save changes to a dashboard's panels and layout from the dashboard view, click
**Save**.

## Copy a panel to a dashboard

Make a copy of a panel in an existing dashboard or copy a panel to another dashboard by
using one of the these methods:

To add a panel to a notebook, see
[Add items to your notebook](/navigate/notebooks#add-items-to-your-notebook).

To copy a panel to any dashboard using keyboard shortcuts:

1. Open a dashboard.
2. Click the panel you want to copy.
3. Press `Control+C` (`Command+C` on macOS) to copy the panel to
   the clipboard.
4. Open the target dashboard.
5. Click a panel group to select it. If you don't select a group, the panel pastes
   into the last panel group on the page.
6. Press `Control+V` (`Command+V` on macOS) to paste the panel into
   the dashboard. The panel uses the same configuration as it had on the origin
   dashboard.
7. Click **Save** to save the updated dashboard.

To customize your panel while copying a panel to another dashboard, or a new,
personal dashboard:

1. Open a dashboard.
2. On the panel you want to copy, click the <Icon icon="ellipsis-vertical" />three dots
   icon, and then select **Add to dashboard**.
3. In the **Add a dashboard** dialog, enter the following information:
   * A **Panel name**: The dialog populates with the name of the panel you selected.
     Keep this name, or update the field with a new name.
   * Choose an option for **Select an existing dashboard**:

     <Tabs>
       <Tab title="Leave blank" id="copy-panel-leave-blank">
         Leave **Select an existing dashboard** blank to create a new dashboard. The
         dashboard will be added to your personal collection. If you don't have a
         personal collection, one will be created for you.

         1. Click **Create new dashboard**.
         2. The dialog updates with the following fields:
            * **Dashboard name**: The name defaults to `[Quick Dash]: USER_ID - TIMESTAMP`
              where:
              * *`USER_ID`* is your username.
              * *`TIMESTAMP`* is the date and time you clicked the **Add to dashboard**
                button.
            * **Owner**: Optional: Select an owner for the dashboard.
       </Tab>

       <Tab title="Select an existing dashboard" id="copy-panel-select-existing">
         If you select an existing dashboard, the panel is added to the selected
         dashboard.

         Select a **Panel group** to add the panel to.
       </Tab>
     </Tabs>
4. Optional: Update the **Preview**.
5. Click **Add to dashboard** when you're finished.
6. The message `Panel successfully added to dashboard. Go to dashboard.` displays.
   Click the **Go to dashboard** link to view the dashboard.

## Undo the previous action

When you paste a copied panel or delete a panel, a message confirming your action
displays in the bottom-left corner of the dashboard. This message summarizes the
action you took (for example, `Pasted panel: test 2`). Next to the message is an
**Undo <Icon icon="undo" />** link. If you have unintentionally pasted or
deleted the panel, click the **Undo** link to restore the panel to its previous state.