Service Level Agreements hold organizations accountable to their users. A high level of service uptime is desirable, but a small margin for error is typically acceptable if established in the service agreement between an organization and its users. The Uptime.com SLA Report allows your organization to report on specific checks or systems for the purposes of SLA accountability.
This article will walk you through the creation of an SLA Report, presenting multiple views to help you determine the best method for your organization to report on Service Level Agreements.
- How to Find SLA Reports
- Downloading an Existing SLA Report
- Editing an Existing SLA Report, or Creating a New SLA Report
- Default Date Range
- Uptime or Response Time Section
- Selecting Checks
- Setting Check SLA Values
- Displaying and Sorting Checks
- Uptime Section
- Response Time Section
- Viewing and Reading Reports
- Filtering Dates
- Drill Down Into Individual Checks
- Uptime and Response Time Columns in SLA Reporting
- Color Coded SLA Values
- Response Time SLA Achieved
- Finalizing SLA Reports
To create your first SLA report, click Reports>SLA Reports, then click New SLA Report. By default, all Uptime.com users can view any SLA report, but only users with View & Modify permissions can edit or create new reports.
To edit an existing report, locate the report you wish to edit and click Actions>Edit. You can edit your SLA report to define the checks (either by tag or individually) that your report will display, as well as the logic that determines how the report is rendered.
To delete an SLA report, click Actions>Delete.
To download a PDF or XLS copy of your SLA report, click Reports>SLA Reports, locate the report and then click Actions> Download PDF, or Download XLS.
To create a report, you need to complete several required attributes, including the report’s Name, a Default Date Range, as well as an Uptime and/or Response time section(s). You may add checks at report creation, or at any point by editing a report.
SLA Reports do not have a limit of checks, but some important notes follow:
- You may not run a report containing response time datapoints if the date range is > 31 days.
- When the date range is <=31 days, and checks selected are less than or equal to 100, hourly datapoints are used to calculate the SLA values. This is the recommended limit of checks and reporting range to ensure the most accurate response time SLA calculations.
- If the date range is <=31 days and more than 100 checks are selected, only aggregated daily response time datapoints will be used in response time SLA calculations. This is less accurate than the above method.
- If the date range is more than 90 days ago, which is the limit for hourly data retention, only daily response time datapoints will be used in response time SLA calculations.
The Default Date Range represents the time period an SLA report will display when viewed. Use this setting to designate reporting specific to a set time period, such as last week or last month.
Tip: You may find your team will need multiple reports for various time periods. For example, executives may want a 30 day review, but your team may benefit from the previous 7 days’ worth of reporting.
Your choices are:
- Today: SLA report from 12:01 AM local time up to the present moment
- Yesterday: Previous day’s 24 hour period, from 12:01 AM local time up to 12:00 AM
- Last 7 days: Beginning from the current date, report on the previous 7 days
- Last 30 days: Beginning from the current date, report on the previous 30 days
- This Week: Beginning on Monday and ending on the present day
- Last Week: Monday to Sunday
- This Month: From first of the current month to present day
- Last Month: From the first day to the last day of the previous month
Sections determine what the end user sees in the generated SLA report. When you toggle on Uptime, the user will see SLA reporting measured against service uptime. When you toggle on Response Time, the user will see SLA reporting measured against target response time.
It is possible to run a report containing both of these sections, which will appear in separate tabs.
Auto-Select Checks by Tag
Include all present or future checks with any of the selected tags. Learn more about creating and managing tags to group checks together.
Manually Select Checks
Include checks from a list you create from a pool of all checks within your account.
Tip: You can manually select checks in addition to the checks you’ve selected by tag.
To set a check’s SLA value, target SLA percentage and/or response time, click Monitoring>Checks. Next, either create a check or locate an existing check. Click Actions>Edit Check and navigate to the Advanced tab. Here, you can set:
- Target SLA %
- Target Response Time SLA (secs)
Every check has a default Target SLA % value of 99.00%).
Target Response Time SLA default values differ per check:
- Transaction checks have a default value of 32.0 seconds
- API checks have a default value of 4.0 seconds
- Ping ICMP checks have a default value of 1.2 seconds
- NTP checks have a default value of 1.0 seconds
- Heartbeat and Webhook checks require you to configure a Target Response Time SLA
- And all other checks have a default value of 2.2 seconds
Whether you select to show sections based on Uptime, Response time, or both, you can determine check displaying and sorting logic as follows:
- Show Target SLA Values on Report: Show the uptime target SLA values that a specific check is to meet. This is a default value of the report.
- Show Only Checks with Downtime: Display only checks that have experienced downtime for the Date Range selected.
- Show Only Checks Below SLA: Exclude any checks that meet or exceed their target uptime SLA such that only checks that are below the target SLA are shown.
- Sort By Lowest Uptime: From top to bottom, display checks with the lowest uptime percentage first.
- Sort By Difference from Check SLA: From top to bottom, sort checks by their mathematical difference from their target SLA. (Secondary sort order is alphabetical)
- Show Target SLA Values on Report: Show the response time target SLA value that a specific check is to meet. This is a default value of the report.
- Show Only Slowest Checks: For a given Date Range, show only the 5 slowest performers of the checks selected.
- Show Only Checks Below SLA: Exclude any checks that meet or exceed their target response time SLA such that only checks that are below the target SLA are shown.
- Sort By Slowest Response Time: From top to bottom, display checks with the slowest response time first.
- Sort By Difference from Check SLA: From top to bottom, sort checks by their mathematical difference from their target response time. (Secondary sort order is alphabetical)
The Uptime.com SLA Report always uses response time datapoints of either 1-hour or 1-day as noted. After a period of 90 days, only a daily response time average is visible based on account alert retention. The same is true if an SLA report contains more than 100 checks.
If accuracy is a concern, Uptime.com recommends users consider splitting SLA reports into smaller groupings.
To view a report, click Reports, locate an existing SLA report and click Actions>View Report or click on the name of the report. This section walks through various aspects of your report, and how changes made in your report’s edit screen affect what users will see.
A typical SLA report Metric Summary will include:
- Uptime for Period, expressed as a percentage
- Total number of Outages
- Average Target SLA for all checks
- Links to Check Reports for drill-down and deeper review
Users may filter dates for a specific SLA report regardless of the Default Date Range. Click the calendar box at the top right-hand side and select the date range for the SLA report you wish to view.
Click an individual check to view that check’s report screen.
Here, you will find the checks within your report, and how their uptime compares to the target SLA value. Note: Average Target SLA value in this section is an average for all checks. This statistic is separate from an individual check’s SLA value.
Certain checks can be excluded from several fields (such as Uptime %, downtime, or Response Time). These omitted checks include checks that run only once-per day, and checks that do not collect response time statistics. Paused checks will also be omitted, as well as checks where Include in Metrics is unticked.
Note: Look for the “i” icon (which provides additional information about checks that are excluded from global uptime and downtime calculations).
Check types in the Response Time Section are grouped together to reflect averages of each check type (e.g. HTTP, API etc.).
Target SLA %
If the SLA/Uptime value is above the Target SLA % by 50% or greater, the numbers will appear green. For example, if Target SLA % is 90%, and the check is 95%, the numbers will be green.
If the SLA/Uptime value is between 50% above the target and the target SLA %, the value will change to Orange color. Using the 90% SLA value, the check will switch to Orange when the SLA value reaches between 94.99% and 90%.
If the check falls below its SLA target, 90% in our example, the value switches to red.
Response Time SLA (secs)
If the Response Time SLA is below 70% of the Target Response Time the numbers will be Green. For example, if the Target Response Time SLA is 10 seconds and the check is between 0s-7s the numbers will appear Green.
If the Response Time SLA is between 70%-100% of the Target Response Time, the numbers will be Orange. In our example of a 10s target this range is from 7.01s-10s.
If the check fails above the 10s Target Response Time, the value switches to Red.
This value refers to the proportion of time that the check performed better/faster than the target SLA number or alternatively, the proportion of time SLA was met for a given period.
SLA values are based on hourly datapoints to calculate response time SLA unless otherwise noted. Such as:
- Over 100 checks
- Over 90 days old daily data is used
To view your SLA Report usage, click Billing>Plan Usage to view your plan usage page. SLA Reports will soon be shareable, with more options. Please submit any feedback about this feature, or reach out to email@example.com for assistance.