About the Edge Observer page

The Edge Observer page allows you to view metrics for Fastly Deliver and Compute services via system-generated overviews and customizable dashboards. This page also gives you access to the Compute@Edge log tailing feature, which allows you to view custom log messages from your Compute@Edge application.

Before you begin

Be sure you know how to access the web interface controls before learning about the information you'll encounter here.

About the Edge Observer page

When you access the Edge Observer page, the Account Summary dashboard appears by default. This dashboard provides a high-level overview of all your services. You can view additional dashboards that focus on a single service by clicking the System Dashboards and Custom Dashboards links.

A series of controls appear to the right of and below the dashboard name. These controls allow you to refine the displayed metrics on any dashboard. Specifically:

• the Options menu allows you add event markers to the charts. Event markers are vertical lines that signify when a new version of the service was activated.
• the Service menu allows you to specify the service that the metrics are drawn from.
• the Region menu allows you to limit the displayed metrics on any dashboard to a specific region around the world.
• the Live button (which only appears on dashboards for specific services, not the Account Summary dashboard), allows you to change displayed metrics to live data that changes in real time.
• the Time Range menu allows you to change the timeframe over which metrics will be displayed and to change the timezone from Coordinated Universal Time (UTC) to the timezone identified by your browser. By default, the system displays 1 days worth of static data. The selected timeframe and timezone persist as you navigate between dashboards.

Hovering over any part of a graph displays a timestamp indicator that updates itself as you move the cursor.

Account Summary dashboard

The Account Summary dashboard provides a high-level overview of all your services. It displays the following metrics for each service in a table:

• Requests: the number of requests Fastly receives for your service over time.
• Errors: the number of cache errors.
• Bandwidth: the number of bytes delivered from Fastly's servers to your website's visitors.

The table of all services allows you to:

• sort the services in the summary table by clicking the column header that you want to sort by (e.g., clicking the Requests header will list the service that received the highest number of requests first and the service that received the fewest number of requests last).
• view additional metrics for a service by clicking the See more details link in the row of the relevant service. The Service Overview dashboard appears.
• click the star icon ⭐ next to services that you want to favorite. The table lists services with stars first.

Service Overview dashboard

The Service Overview dashboard displays different information depending on if you are viewing a Deliver or Compute service.

Service Overview dashboard for a Deliver service

When a Deliver service is selected from the Service menu, the Service Overview dashboard displays the following metrics:

• Cache Hit Ratio: the percentage of cache hits to all cacheable content over time.
• Latest Cache Hit Ratio: the percentage of cache hits to all cacheable content.
• Requests: the number of requests by cache response type.
• All Status Codes: the number of responses sent by status group.
• Errors: the number of cache errors.
• Misses: the number of cache misses.
• Object Sizes: the number of objects served by payload size.
• Hit Time: the total amount of time spent processing cache hits.
• Miss Time: the total amount of time spent processing cache misses.
• Bandwidth: the total bytes delivered.
• Log Lines: the number of log lines sent.
• Log Bandwidth: the total log bytes sent.

Service Overview dashboard for a Compute service

When a Compute service is selected from the Service menu, the Service Overview dashboard has two tabs: Metrics and Logs.

The Metrics tab provides the following details for the selected Compute service:

• Requests: the total number of requests that were received for your service by Fastly.
• Response status codes: the total number of response status codes by category delivered by Compute@Edge.
• Total resource limits exceeded: the number of times a guest exceeded its resource limit, includes heap, stack, globals, and code execution timeout.
• Memory Resource Limits Exceeded: the number of times a guest exceeded its heap, stack, and global limits.
• Backend Requests Errors: the number of backend request errors, including timeouts.
• Edge Code Errors: the number of times a service experienced an edge code error.
• Guest Runtime Errors: the number of times a service experienced a guest runtime error.
• Bytes Received by Compute@Edge: the total bytes received by Compute@Edge.
• Bytes Transferred to Client: the total bytes sent from Compute@Edge to end user.
• Bytes Transferred to Backend: the total bytes sent to backends (origins) by Compute@Edge.
• Bytes Transferred from Backend: the total bytes received from backends (origins) by Compute@Edge.
• Total Wall Clock Time: the total, actual amount of time used to process your requests, including active CPU time.
• CPU Time: the amount of active CPU time used to process your requests.
• RAM Usage: the amount of RAM used for your service by Fastly (in bytes).
• Logs: the number of logs sent to your endpoints from Fastly.
• Log Bandwidth: the total bandwidth size of the logs sent to your endpoints from Fastly.
• Fanout (Published Messages): messages sent to end users and messages received from the publish API endpoint.

The Logs tab gives you access to the Compute@Edge Log Tailing feature. Log Tailing streams your service's log output to the Fastly web interface so you can respond quickly when debugging your application without setting up a third-party logging tool. Any output sent to the standard output stream (stdout) and the standard error stream (stderr) will appear on the Logs page, as well as any runtime errors encountered by the application.

Specifically, the Logs tab displays the following information for the selected Compute service:

• Date (UTC): a UTC timestamp of when the request occurred.
• Request ID: a unique identifier applied to the request.
• Event: the event type, either stdout or stderr.
• Message: the message included in the request.

Click the Pause button to pause the live stream of Compute logs. Click the Live button to resume the live stream.

Origin Inspector dashboard

The Origin Inspector dashboard displays the following metrics:

• Responses by Origin: the number of responses processed by each origin host.
• Response body bytes by Origin: the number of response body bytes returned by each origin host.
• Response header bytes by Origin: the number of response header bytes returned by each origin host.
• 2xx by Origin: the number of 2xx type (success) HTTP response status codes returned by each origin host.
• 3xx by Origin: the number of 3xx type (redirection) HTTP response status codes returned by each origin host.
• 4xx by Origin: the number of 4xx type (client error) HTTP response status codes returned by each origin host.
• 5xx by Origin: the number of 5xx type (server error) HTTP response status codes returned by each origin host.
• Origin Latency: a histogram showing the distribution of origin latency times. This tells you how quickly your origin is responding to Fastly.

Domain Inspector dashboard

The Domain Inspector dashboard displays the following metrics:

• Edge by Domain: the number of requests processed at the edge for each domain.
• Response body bytes by Domain: the number of response body bytes delivered from the edge for each domain.
• Response header bytes by Domain: the number of response header bytes delivered from the edge for each domain.
• 2xx by Domain: the number of 2xx type (success) HTTP response codes delivered for each domain.
• 3xx by Domain: the number of 3xx type (redirection) HTTP response status codes delivered for each domain.
• 4xx by Domain: the number of 4xx type (client error) HTTP response status codes delivered for each domain.
• 5xx by Domain: the number of 5xx type (server error) HTTP response status codes delivered for each domain.

Image Optimizer dashboard

The Image Optimizer dashboard displays the following metrics:

• Image Optimizer total responses: the total number of responses that came from the Fastly Image Optimizer service.
• Image Optimizer shield responses: the number of responses that came from the Fastly Image Optimizer service via a shield.
• Image Optimizer total bytes delivered: the total bytes delivered from the Fastly Image Optimizer service, including shield traffic.
• Image Optimizer shield bytes delivered: the total bytes delivered via a shield from the Fastly Image Optimizer service.
• Image Optimizer video responses: the number of video responses that came from the Fastly Image Optimizer service.
• Image Optimizer video frames: the number of video frames that came from the Fastly Image Optimizer service. A video frame is an individual image within a sequence of video.
• Image Optimizer video bytes delivered: the total bytes of video delivered from the Fastly Image Optimizer service.
• Image Optimizer video shield responses: the number of video responses delivered via a shield that came from the Fastly Image Optimizer service.
• Image Optimizer video shield bytes delivered: the total bytes of video delivered via a shield from the Fastly Image Optimizer service.
• Image Optimizer video shield frames: the number of video frames delivered via a shield that came from the Fastly Image Optimizer service. A video frame is an individual image within a sequence of video.

Fanout dashboard

The Fanout dashboard displays the following metrics:

• Fanout (Connection Time): the total duration of Fanout connections with end users.
• Fanout (Published Messages): the total number of messages received from the publish API and sent to end users.
• Fanout (Bytes transferred to client): the total bandwidth transferred by Fastly for your service when serving requests.
• Fanout (Bytes transferred from client): the total bandwidth transferred by Fastly for your service when receiving requests.
• Fanout (Bytes transferred to backend): the total bandwidth transferred by Fastly to your backend when receiving requests.
• Fanout (Bytes transferred from backend): the total bandwidth transferred to Fastly from your backend when serving requests.

WebSockets dashboard

The WebSockets dashboard displays the following metrics:

• WebSockets connection time: the total duration of passthrough WebSocket connections with end users.
• WebSockets bytes transferred to client: the total bandwidth transferred by Fastly for your service when using requests.
• WebSockets bytes transferred from client: the total bandwidth transferred by Fastly for your service when receiving requests.
• WebSockets bytes transferred to origin: the total bandwidth transferred by Fastly to your backend when receiving requests.
• WebSockets bytes transferred from origin: the total bandwidth transferred to Fastly from your backend when serving requests.

About data latency

When working with dashboards and the metrics they display, keep the following in mind about data:

• We do not report data for time periods that have not yet ended.
• Data with day resolution is bucketed based on UTC days and each day's data becomes available around 2am the following day.
• Data with hour resolution buckets becomes available approximately 15 minutes after the end of each hour.
• Data with minute resolution buckets usually becomes available two minutes after the end of the minute, but can take up to 15 minutes to appear.

If your use case requires data closer to real-time, consider using the generally available real-time API instead.

What's next