Skip to main content
Sumo Logic

Service Map and Dashboards

The Service Map is a high-level view of your application environment, automatically built from tracing data in real time as it arrives in Sumo Logic. This gives you a greater understanding of your application architecture and dependencies between monitored microservices.

This guide provides information on the following:

Service Map

To open Service Map click + New and select Service Map.

open service map.png

The Service Map provides the following features:

  • Zoom in and out on the map.
  • Hover over a service to see its connections to other services and the last 15 minutes of activity in terms of latency, requests, and errors.
  • Click on a service to open the Entity Inspector with the ability to drill down to traces, metrics, and the service dashboard.
  • Filter by application, if your tracing data has the application=[app-name] tag.
  • Search for a service by name.

service map april 2021.png

Color of services

The color of services is based on automatic (default) or manual thresholds you can configure in the settings menu. Red indicates an anomaly and blue indicates normal activity.

Shape of services

Remote services, like databases or external calls, automatically detected in client traffic, even without actual direct instrumentation, are visualized using the cloud icons instead of circles.

service circles.png


The Service Map shows activity between services and their presence from the last 72 hours, however, if the service or connection was inactive for more than one hour, it is displayed with dashed lines.

Size of services

The size of services on the map is based on how active they are, where large circles are more active compared to small circles that are less active.  


Settings are user-specific and preserved in your browser's Local Storage, so each user can have their own specific settings to view Service Map health indicators. The settings menu opens by clicking the gears icon in the top-right corner of the map.
service map gear icon for settings.png

Anomaly Detection

Automatic anomaly detection is enabled by default. A service is displayed in a red circle if at least one of the KPIs (latency, errors, requests) has an anomaly (a spike for latency and errors or dip in the case of requests) in the last 15 minutes AND the anomaly was not present 7 days ago at the same time. In the settings menu, this default option is called Worst Case. You can explicitly choose one of the KPIs instead.

The standard metric outlier operator is used to detect this. You can adjust the sensitivity and learning window of the outlier detection in the settings menu to make the logic more or less dependent on occasional spikes. 

auto service map settings april 2021.png

Manual anomaly detection settings allow you to specify different detection settings for each KPI. Click the toggle switch to activate these settings on and off. You can explicitly toggle each KPI as desired.

service map manual detection settings.png

Application Service Dashboards

Service and Application dashboards are available as two Explore hierarchies. You can access them in the Explore By drop down menu.

The Application Service View groups services into higher-level applications based on the application=[app-name] custom tag that you may add to your tracing data if you want to leverage it fully. If not present, all services will belong to a "default" application. See how to modify the attributes of your spans. The third level shows the top 50 most active operations executed on the service.

The Service Application View takes a contrary approach, displaying services by top level and breaking down their health by application. This view is especially useful for shared services that support more than one application. The third level shows the top 50 most active operations performed on the selected service and application.

Service Dashboards from traces.png

Services need to be active in the last 15 minutes to appear on the list. Recent inactivity will result in a grayed-out list entry.

Tracing metrics retention has default metrics retention.

Each dashboard is a fully customizable set of panels based on automatically generated metrics using tracing data as input. You can get insight into microservice health by looking at stats from:

  • Latency: (real-time average) the average time it takes for entry spans for that service to complete. Apart from average you can also select p99, p95, p90, p50 options in service_latency_type filter to view other percentile aggregations.
  • Requests: (real-time counter) the number of entry spans reported by the service
  • Errors:  (real-time counter) the number of entry spans for the service that finished with an error

By clicking on any data point on the chart, just like for any other Sumo Logic dashboard, you can view the side panel's Infrastructure tab to drill down to related metrics or traces for the selected service.

Explore coffee app to drill down on infrastructure tab.png

Install the App

The Tracing - Application Services Health App is automatically installed for all users of your organization once Sumo Logic detects OT compatible tracing data coming from your instrumented services. The content is placed in Sumo Logic Tracing - default dashboards inside the Admin Recommended folder and is automatically available for all users in the organization.

To install the app, do the following:

  1. From the App Catalog, search for and select the Tracing - Application Services Health.
    If you want to see a preview of the dashboards included with the app before installing, click Preview Dashboards.

  2. To install the app, click the Add to Library button and complete the following fields.

  • App Name. You can retain the existing name, or enter a name of your choice for the app.

  • Advanced. Select the Location in Library (the default is the Personal folder in the library), or click New Folder to add a new folder.

Trace app two.png

  1. Click Add to Library.

Once an app is installed, it will appear in your Personal folder or the folder you specified. From here, you can share it with your organization.

Dashboard Panels

Add Service Map and Trace List panels to explore your application environment and review all traces directly through your Dashboard. Duplicate or add multiple panels with different filtering or queries to refine views and support your organization. 

See the Dashboard (New) guide for additional information and options to create panels, configure filters, create and filter with template variables from dashboard headers, and more.

Service Map panel

The Service Map panel displays a service map for filtered applications and services, with the ability to explore the environment.

To add a Service Map panel to your Dashboard:

  1. Open a Dashboard or create a new Dashboard.

  2. Click Add Panel and select Service Map. Or select the Service Map option on a new Dashboard.
    A panel configuration page opens.

  3. Select from the drop-down menus to filter the Service Map by the following:
    • Application if your tracing data has the application=[app-name] tag
    • Service 
    • To pass the variables from dashboard filters, set application={{application}} and/or service={{service}}
  4. Do not set the time. Service Map always shows last 72h of data.

  5. The Chart Type is set to Graph.

  6. Click the General tab to edit the Panel Details. Enter a name for the panel, set a Title Font Size, and add a short Description.

  7. Click Add to Dashboard. The panel loads in your Dashboard to review your applications and services according to filtering.

Trace List panel

The Trace List panel displays the Traces page table to give at-a-glance tracking for traces. To view deeper details, click a trace from the panel to open the Trace View

To add a Trace List panel to your Dashboard:

  1. Open a Dashboard or create a new Dashboard.

  2. Click Add Panel and select Trace List. Or select the Trace List option on a new Dashboard.
    A panel configuration page opens.

  3. Configure a Trace query to search for desired set of traces. 

  4. Select a time range or create a custom range for the panel. You can set this when creating or at any time when viewing the Dashboard panel.

  5. The Chart Type is set to Table.

  6. Enter the Rows Per Page for the panel, between 5 to 100. The default amount is 15. The panel automatically paginates traces to browse through and view all traces.

  7. Select the Table columns of trace data to load in the panel:

Column name Example value Description
Trace ID ffaf2f69ee8ad0c1 The unique identifier of the trace.
Root Service api The service that started the trace.
Started At 07/27/2020 09:01:04.533 When the trace started.
Duration 12.582 ms The amount of time the trace spans. 
Number of spans 35 A trace consists of spans. This number tells you how many spans are in the trace.
Duration Breakdown breakdown.png Each color indicates a service. The colors assigned to services are always the same on your account. You can change the color in the span summary tab after clicking on the individual span in trace view.

Hover over to view a percentage breakdown of how long each span covers in the trace.

Number of errors 0 The number of errors in the trace.
Status 200 The HTTP status code of the trace.
  1. Click the General tab to edit the Panel Details. Enter a name for the panel, set a Title Font Size, and add a short Description.

  2. Click Add to Dashboard. The panel loads in your Dashboard to review all traces according to the query.