Services

Define and manage services in ilert to reflect availability status and impact across incidents and status pages.

Users can subscribe to services to receive notifications about status updates that affect specific business capabilities.

A service has any of the following states:

Status	Description
Operational	All aspects of the service are working as expected.
Under maintenance	The service is under maintenance.
Degraded	The service is working but is impacted in a minor way, e.g. its performance is slower than usual.
Partial outage	The service is not working for a subset of customers, e.g. a certain region.
Major outage	The service is not available.

The status of a service can be updated when you post a status update, or independently on the service itself.

Service topology

Services can depend on one another. Connect a telemetry source and ilert reconstructs your live service topology from OpenTelemetry traces — visualizing dependencies, blast radius and likely root cause on the Service Health map.

Service status changes and notifications

Updating a service's status directly on the service does not notify subscribers about the change. Subscribers are notified only when a status update is posted with affected services they are subscribed to.

Create a service

To create a new service, navigate to the Services page, click on the Create new service button.

Each service carries its own 60-day uptime history, and users can subscribe per service.

Give the service a name that is descriptive to the users of the service and click on Create new service.

A service needs only a name to start with.

The service is now ready to be used in incidents or included in status pages.

Automation with alert sources

You can automatically set the status of a service and post status updates using alert actions that are triggered by your alert sources.

To configure an automation

• in the sidebar, go to Alerting → Alert sources

Pick the alert source whose events should drive the service status.

• choose an alert source and navigate to the Alert actions tab
• click on Create new alert action
• select the ilert incidents action type and click Next

Selecting the ilert incidents action type

Once you have configured the automation, any alert created on the chosen alert source will run the action and—depending on how you configure it—set service status or post a status update. Make sure that the trigger mode is set to Automatic for the action to run on every alert:

Setting the trigger mode to Automatic

Service uptime

How to hide historical service uptime

By default, a service's historical uptime is shown and you can choose to hide service uptime on the status page. If you'd like to globally disable historical uptime, thus preventing any status page from displaying the historical uptime, you can do so in the service settings:

1. Go the services page and click on the service for which you want to hide the uptime
2. In the Settings tab, check the option Never show historical uptime
3. Click on Save

How is the uptime of a service calculated?

The uptime percentage of a service is calculated based on the status of the service over a given period. ilert displays the uptime over a period of up to 90 days. We consider the states Operational, Degraded and Under maintenance as uptime. The states Major outage and Partial outage are counted as downtime, where partial outage minutes only count 30% as much as major outages.

Over a period t, the uptime percentage is calculated according the following formula:

$$u=1-{m+(p*0.3) \above{1pt} t}$$

where

• u is the uptime percentage
• m is the major outage in minutes
• p is the partial outage in minutes
• t is the period in minutes

Example: During a period of 24 hours, a service was 5 minutes in the status Major outage and 30 minutes in the status Partial outage. The uptime of the service for that 24 hours period is calculated as follows:

$$u=1-{5+(30*0.3) \above{1pt} 1440}=99.03 \%$$

Outages that last less than a minute will be ignored and won't show up as downtime in the uptime graph.

How is the color of each day in the uptime graph determined?

Each day in the uptime graph is color-coded based on the total outage minutes of that day. The outage minutes are calculated using the same weighted formula as above, where partial outage minutes count 30% as much as major outages.

The color ranges from green (no or minimal downtime) to red (significant downtime):

Total outage minutes	Color
0 minutes	Green
> 0 – 6 minutes	Light green
> 6 – 9 minutes	Lime
9 – 12 minutes	Light yellow
12 – 25 minutes	Yellow
25 – 30 minutes	Light orange
30 – 40 minutes	Orange
40 – 45 minutes	Dark orange
45 – 50 minutes	Light red
50 – 60 minutes	Red
60+ minutes	Dark red

Days in the future as well as days before the service was created (where no uptime data exists) are displayed in grey.

The transition from green to red is gradual — even a few minutes of outage will shift the color slightly, giving you a quick visual indication of service health at a glance.

Add / remove subscribers to a service

Subscribers will automatically receive notifications when the Send notifications option is selected on a status update. There are two ways to subscribe users to a service

1. Users can add / remove themselves as subscribers
2. A user with Responder privileges can manually subscribe users or entire teams to a service

To self-subscribe to a service, go the Services page and click on Subscribe. To unsubscribe, click on the Unsubscribe button.

Team subscriptions

Note that an unsubscription takes precedence over a team subscription. That is, if you are subscribed to a service, because you are a member of a team that is subscribed to the service, and if you unsubscribe yourself from that service, you won't receive any further updates from that service.

To add or remove users and teams as subscribers from a service, go to the Subscribers tab from the service.