Help Center

  1. StackPath Help
  2. Monitors and Alerts
  3. Monitors

Create and Manage Monitors

Avatar
George Alpizar
Updated:

Overview

You can use this document to learn how to:

  • Create and update monitors
  • Review and understand monitor results

A monitor allows you to observe the performance of your public-facing websites and services from a global perspective. After a successful sync, every 5 minutes a monitor will initiate a connection and collect metrics for the monitored resource. The collected metrics can help you detect website performance issues.

Create a monitor

You can create an HTTP or TCP monitor. 

An HTTP monitor sends a request on a port and expects a web server to respond. The request replicates a user that browses to a URL to measure the full-page load time. Custom headers, HTTP basic auth, and body of content can be sent. 

A TCP monitor opens a TCP socket on a specified port to check that a connection can be made and content can be read from the socket. A custom string can be sent with the connection before the read operation takes place. 

Option 1: Create an HTTP monitor

  1. In the StackPath Control Portal, in the left-side navigation, click Monitoring.
  2. Click Create Monitor. 
  3. In Name, enter a descriptive name. 
    • This name will display in the Monitors table within the Monitoring screen. 
  4. In Protocol, select HTTP.
  5. In URL, select HTTP or HTTPS, and then enter the URL or IP address to monitor. 
  6. In Method, select the response method. 
  7. In IP Version, select the IP version that will be used to connect to the monitored resource.  
  8. In Location, select the location where the monitored URL or IP address will be monitored from. 
    • You can select multiple locations. 
    • Each selected location will send a separate request to the monitored URL or IP address and will have separate monitoring data. 
    • While you can select China as a location, there may be response times or connection issues due to local internet regulations. As a result, if applicable, do not add China to your list of locations.  
    • To ensure that monitors can connect to your monitored resource, StackPath recommends that you whitelist the monitoring IP addresses. To view the list of IP addresses, please see View a Monitoring Location's IP Address and Service Provider.
  9. By default, Frequency will send out a request every 5 minutes.  
  10. In Timeout, enter how long the monitor will attempt to make a connection to the monitored URL or IP address.
    • If a timeout is reached, then the website will be considered down for the specified location. 
    • By default, the timeout is set to 10 seconds.
    • The maximum timeout is 20 seconds.
  11. (Optional) If you want StackPath to check that the monitored site's SSL certificate has not expired nor been revoked, then mark Strict SSL Verification.
    • If there certificate has expired or been revoked, then the monitored site will be considered down.
    • This option is useful if you want to be notified when your SSL certificate needs to be renewed. 
  12. (Optional) If you want the monitors to use an authentication method, then in Authentication, select Basic or Digest, enter the credentials, and then enter a public certificate and private key. 
  13. (Optional) If you want to POST data in the request body as part of the HTTP check request, then under Request Body Content, enter the body content to send with the check. 
    • You can only use this option if you previously set the Method to Post. 
    • This option is useful if you want to monitor an endpoint for availability and responsiveness that only accepts POST requests.
    • You can set a header to complement the body content. By default, the Content-Type header to application/x-www-form-urlencoded is automatically set when the body content is configured and a Content-Type header is not configured. You can add another value for your Content-Type header, such as application/json or application/xml.
  14. Click Create Monitor. 

After you create a monitor, StackPath recommends that you create an alert to complement the monitor. To learn more, see Create and Manage Alerts.

Option 2: Create a TCP monitor  

  1. In the StackPath Control Portal, in the left-side navigation, click Monitoring.
  2. Click Create Monitor. 
  3. In Name, enter a descriptive name. 
    • This name will display in the Monitors table within the Monitoring screen. 
  4. In Protocol, select TCP.
  5. In Host, enter the URL or IP address to monitor. 
  6. In Port, enter the port that the monitor will create a TCP connection on. 
  7. In IP Version, select the IP version that will be used to connect to the monitored resource.  
  8. In Location, select the location where the monitored URL or IP address will be monitored from. 
    • You can select multiple locations. 
    • Each selected location will send a separate request to the monitored URL or IP address and will have separate monitoring data. 
    • While you can select China as a location, there may be response times or connection issues due to local internet regulations. As a result, if applicable, StackPath recommends that you do not add China to your list of locations.  
    • To ensure that monitors can connect to your monitored resource, StackPath recommends that you whitelist the monitoring IP addresses. To view the list of IP addresses, please see View a Monitoring Location's IP Address and Service Provider.
  9. By default, Frequency will send out a request every 5 minutes.  
  10. In Timeout, enter how long the monitor will attempt to make a connection to the monitored URL or IP address.
    • If a timeout is reached, then the website will be considered down for the specified location. 
    • By default, the timeout is set to 10 seconds.
    • The maximum timeout is 20 seconds.
  11. In Socket Message, enter a data string to send when the socket opens. 
  12. Click Create Monitor. 

After you create a monitor, StackPath recommends that you create an alert to complement the monitor. To learn more, see Create and Manage Alerts.

Add or remove locations for an existing monitor

  1. In the StackPath Control Portal, in the left-side navigation menu, click Monitoring. 
  2. Locate and select the desired monitor. 
  3. In the left-side navigation menu, click Locations.
  4. Click Monitoring Locations, and in the menu that expands, slide to on or off for a location. 
    • Changes are saved automatically and will display in the same screen under PoPs Monitoring this Domain. 

View existing monitors and understand monitor status

  1. In the StackPath Control Portal, in the left-side navigation, click Monitoring.
  2. Review the table of existing monitors: 
    Column Description
    Name This column displays the friendly name of the monitor. 
    Host This column displays the URL of the monitored resource. 
    Avg. Response Time This column displays the average response time to complete a full check. 
    Uptime  
    Status

    This column displays if the monitored site is up, down, or slow.

    A monitor is Up when StackPath receives any response, which can include error status codes, such as a 5xx error message.

    A monitor is Down when StackPath cannot connect to the monitored resource, the URL/IP address resolves, or if there is another error.

    Errors include name server lookup failures, timeouts, or SSL certificate errors. 

    A connection is attempted again to verify the state.

    Timeouts are capped at 20 seconds, regardless of what has been configured in the portal.

    A Down status will not be recorded for error pages or HTTP status codes. For example, a 500 internal server error is considered Up because the monitor received a response from the monitored resource.

    A monitor is Slow when any response time is greater than 500 milliseconds.
    Locations This column displays the number of locations where the monitored URL or IP address is monitored from. 
    Check Type  This column display if the monitor is HTTP or TCP based. 
    Method

    This column displays the response method (POST or GET) for HTTP monitors.

    For TCP monitors, this column will be blank.

View and understand results for a specific monitor

You can use these instructions to view detailed information for a monitor, specifically response time metrics for each phase of the request. This information can help you detect and troubleshoot slow response times. 

  1. In the StackPath Control Portal, in the left-side navigation, click Monitoring.
  2. Locate and select the desired monitor. 

Understand Average Response Time graph 

By default, the Average Response Time graph displays data for Request Components; however, this graph can also display data for Response Time. To update the graph to display response times, in the top, right corner of the graph, click the gear icon, and then select Response Time. 

Request Components graph for Average Response Time

Metric Description
DNS Lookup Time

This data displays the length of time the DNS resolution took to complete. 

This data is measured in seconds. 
TCP Connection Time

This data displays the amount of time the TCP took to connect to the remote host. 

This data is measured in seconds. 
TLS Handshake Time

This data displays the amount of time the TLS handshake negotiation took to complete. 

This data is measured in seconds. 
Server Processing Time

This data displays the amount of time the server took to process the request before the first byte was received. 

This data is measured in seconds. 

Response Time graph for Average Response Time

Metric Description
Average

This column displays the average response time to complete a full check. 
Trending

This column displays an up arrow or down arrow to indicate if the response time is higher or lower than the average time.

The displayed time represents the response time for the last check. 
High / Low

This column displays the highest and lowest values. 

Customize the Average Response Time graph 

Review the following ways to personalize the Average Response Time graphs. 

  • Update displayed time
    • By default, data is displayed from the previous 24 hours. To change the time frame, in the top, right section, click the time frame, and then select a new time frame. 
  • Update displayed metrics
    • By default, all four metrics (DNS Lookup Time, TCP Connection Time, TLS Handshake Time, Server Processing Time) are displayed in the graph. To add or remove displayed metrics, above the graph mark or unmark the metrics; the graph will refresh. 
  • Display response times
    • By default, the Average Response Time graph displays data for Request Components; however, this graph can also display data for Response Time. To update the graph to display response times, in the top, right corner of the graph, click the gear icon, and then select Response Time. 
  • Export data
    • You can export the graph as a PNG, PDF, CSV, or XLS file. You can also directly print the graph. In the right side of the graph, click the horizontal lines, and then select the desired output type. 

Understand the Response Time By PoP graph

By default, data is displayed from the previous 24 hours. To change the time frame, in the top, right section, click the time frame, and then select a new time frame. 

Metric
Description

Location

This column displays the city and country of the PoP location. 

Provider

 This column displays the virtual private server provider that performs the check.  

Average

This column displays the average response time for a completed full check. 

Trending

This column displays an up or down arrow to describe if the last response time

The arrow displays if the last response time is trending up or down compared to the average. The time displayed is the response time of the last check.

Status Code

This column displays the status code received from the last check. 

High / Low

This column displays the highest and lowest values. 

 Graph

This column displays a sparkline chart to display response times. 

You can hover over a specific point in the graph to view a pop-up window with additional information. 

Customize the Response Time by PoP table

Review the following ways to personalize the Response Time By PoP Time table. 

  • Update displayed time
    • By default, data is displayed from the previous 24 hours. To change the time frame, in the top, right section, click the time frame, and then select a new time frame. 
  • Update displayed metrics
    • By default, all providers, countries, cities, and PoP are displayed. To add or remove displayed metrics, above Average Response Time, click Filter, and then mark or unmark the desired data points to display. 
Was this article helpful?

Related articles

Not finding what you need?

95% of questions can be answered using the search tool. This is the quickest way to get a response.

Contact Us