OpsDuty Incidents¶
An incident is the most important resource in OpsDuty. It describes an event that requires attention from one or more users.
Incident States¶
An incident can exist in one of the following states:
- Triggered: An incident is in the triggered state until someone reacts to it. It inherits the escalation policy from the associated service, and OpsDuty begins executing this policy when the incident is triggered. The escalation policy execution stops automatically when the incident transitions to another state.
- Acknowledged: A user can acknowledge an incident, indicating that they have started investigating the issue and are working on a resolution.
- Silenced: The incident is silenced and not actively being worked on.
- Resolved: A resolved incident indicates that the issue is no longer ongoing.
OpsDuty creates incidents based on incoming events. It determines whether to create a new incident or append the event to an existing one based on a deduplication identifier.
Silence¶
An incident can be silenced for a specific period. When silenced, OpsDuty pauses the escalation policy execution. If the incident is not acknowledged or resolved before the silence period ends, it will transition back to the triggered state.
Reassign¶
Incidents can be reassigned to a different escalation policy. When reassigned, OpsDuty replaces the current escalation policy and restarts the process with the newly selected policy.
Direct Paging¶
The direct paging feature allows users to directly notify a user or team without modifying the escalation policy. This is useful for sending notifications directly to an individual or all members of a team.
Attached Incidents¶
Open incidents can be attached to one another. This is helpful when dealing with duplicate incidents or related incidents where maintaining the state of a single root incident is preferred.
Incident Grouping and Deduplication Identifier¶
OpsDuty uses a deduplication identifier to group related events. A group of events becomes an incident in OpsDuty. If no deduplication identifier is provided, OpsDuty assigns one randomly to incoming events.
Incoming events with the same deduplication identifier are appended to any open incident when received and processed by OpsDuty.
Events¶
Incidents are created or updated based on incoming events received by OpsDuty through its configured integrations. Events are constructed from parsed data from receive channels and include the raw data received by OpsDuty.
Key event fields include:
- Summary
- Description
- Source URL
Certain fields also control the incident lifecycle:
- Acknowledge signal: Marks the incident as acknowledged when such an event is added.
- Resolve signal: Marks the incident as resolved when a resolve event is added.
Notes¶
OpsDuty allows users to collect notes to provide additional context throughout the incident lifecycle. These notes are helpful for creating post-mortems or documenting lessons learned after the incident is resolved.
Log¶
Each incident includes a log that records all operations performed by OpsDuty on the incident.