Analyze data from synthetic monitors

The Synthetics app in Kibana both gives you a high-level overview of your service’s availability and allows you to dig into details to diagnose what caused downtime.

The Synthetics Overview tab provides you with a high-level view of all the services you are monitoring to help you quickly diagnose outages and other connectivity issues within your network.

To access this page, go to Observability → Synthetics and make sure you’re on the Overview tab.

This overview includes a snapshot of the current status of all monitors, the number of errors that occurred over the last 6 hours, and the number of alerts over the last 12 hours. All monitors created using projects or using the Synthetics app will be listed below with information about the location, current status, and duration average.

To get started with your analysis in the Overview tab, you can search for monitors or use the filter options including current status (up, down, or disabled), monitor type (for example, journey or HTTP), location, and more.

Then click an individual monitor to see some details in a flyout. From there, you can click Go to monitor to go to an individual monitor’s page to see more details (as described below).

When you go to an individual monitor’s page, you’ll see much more detail about the monitor’s performance over time. The details vary by monitor type, but for every monitor at the top of the page you’ll see:

Each individual monitor’s page has three tabs: Overview, History, and Errors.

The Overview tab has information about the monitor availability, duration, and any errors that have occurred since the monitor was created. The Duration trends chart displays the timing for each check that was performed in the last 30 days. This visualization helps you to gain insights into how quickly requests resolve by the targeted endpoint and gives you a sense of how frequently a host or endpoint was down.

The History tab has information on every time the monitor has run. It includes some high-level stats and a complete list of all test runs. Use the calendar icon () and search bar to filter for runs that occurred in a specific time period.

For browser monitors, you can click on any run in the Test runs list to see the details for that run. Read more about what information is included the in Details for one run section below.

The Errors tab has information on every time a run has failed. It includes a high-level overview of all alerts and a complete list of all run failures. Use the calendar icon () and search bar to filter for runs that occurred in a specific time period.

For browser monitors, you can click on any run in the Error list to open an Error details page that includes most of the same information that is included the in Details for one run section below.

For browser monitors, you can look at results at various levels of granularity:

The journey page on the Overview tab includes:

From here, you can either drill down into:

The page detailing one run for a journey includes more information on each step in the current run and opportunities to compare each step to the same step in previous runs.

At the top of the page, see the Code executed and any Console output for each step. If the step failed, this will also include a Stacktrace tab that you can use to diagnose the cause of errors.

Navigate through each step using Previous and Next .

Scroll down to dig into the steps in this journey run. Click the icon next to the step number to show details. The details include metrics for the step in the current run and the step in the last successful run. Read more about step-level metrics below in Timing and Metrics.

This is particularly useful to compare the metrics for a failed step to the last time it completed successfully when trying to diagnose the reason it failed.

Drill down to see even more details for an individual step by clicking the performance breakdown icon () next to one of the steps. This will take you to the view described below in Details for one step.

After clicking the performance breakdown icon () you’ll see more detail for an individual step.

By default the synthetics library will capture a screenshot for each step regardless of whether the step completed or failed.

Screenshots can be particularly helpful to identify what went wrong when a step fails because of a change to the UI. You can compare the failed step to the last time the step successfully completed.

The Timing visualization shows a breakdown of the time spent in each part of the resource loading process for the step including:

Next to each network timing metric, there’s an icon that indicates whether the value is higher (), lower (), or the same () compared to the median of all runs in the last 24 hours. Hover over the icon to see more details in a tooltip.

This gives you an overview of how much time is spent (and how that time is spent) loading resources. This high-level information may not help you diagnose a problem on its own, but it could act as a signal to look at more granular information in the Network requests section.

The Metrics visualization gives you insight into the performance of the web page visited in the step and what a user would experience when going through the current step. Metrics include:

Next to each metric, there’s an icon that indicates whether the value is higher (), lower (), or the same () compared to all runs over the last 24 hours. Hover over the icon to see more details in a tooltip.

The Object weight visualization shows the cumulative size of downloaded resources by type, and Object count shows the number of individual resources by type.

This provides a different kind of analysis. For example, you might have a large number of JavaScript files, each of which will need a separate download, but they may be collectively small. This could help you identify an opportunity to improve efficiency by combining multiple files into one.

The Network requests visualization is a waterfall chart that shows every request the page made when a user executed it. Each line in the chart represents an HTTP network request and helps you quickly identify what resources are taking the longest to load and in what order they are loading.

The colored bars within each line indicate the time spent per resource. Each color represents a different part of that resource’s loading process (as defined in the Timing section above) and includes the time spent downloading content for specific Multipurpose Internet Mail Extensions (MIME) types: HTML, JS, CSS, Media, Font, XHR, and Other.

Understanding each phase of a request can help you improve your site’s speed by reducing the time spent in each phase.

Without leaving the waterfall chart, you can view data points relating to each resource: resource details, request headers, response headers, and certificate headers. On the waterfall chart, select a resource name, or any part of each row, to display the resource details overlay.

For additional analysis, whether to check the content of a CSS file or to view a specific image, click the icon located beside each resource, to view its content in a new tab.

You can also navigate between steps and checks at the top of the page to view the corresponding waterfall charts.