On-Call & Escalation
SolidPing can route incidents to the right person at the right time using on-call schedules and escalation policies. Together they answer two questions: who is responsible right now? and what happens if they don't respond?
On-Call Schedules
A schedule defines a rotation of users who take turns being on call.
| Setting | Description |
|---|---|
| Name / Slug | Schedule identity |
| Timezone | Local timezone used for handoff calculations (DST-safe) |
| Rotation Type | daily or weekly |
| Handoff Time | Wall-clock time of day the rotation hands off (HH:MM) |
| Handoff Weekday | Day of week for weekly rotations |
| Start At | Anchor point for the first handoff |
| Roster | Ordered list of users in the rotation |
At any moment, the schedule resolves to exactly one on-call user by walking the rotation from the start anchor.
Overrides
Need someone to cover a shift? Add a time-bounded override that temporarily replaces the scheduled user between a start and end time, with an optional reason. Overrides take precedence over the normal rotation.
iCal Feed
Each schedule can publish a private iCal feed, so on-call shifts show up in Google Calendar, Apple Calendar, or Outlook. The feed token can be enabled, disabled, or rotated at any time.
Escalation Policies
An escalation policy is an ordered list of steps. When an incident fires, the policy walks down the steps, waiting between each, until the incident is acknowledged or the policy is exhausted.
Steps and Targets
Each step waits a configurable delay, then notifies one or more targets in parallel. A target can be:
| Target Type | Notifies |
|---|---|
user | A specific user |
schedule | Whoever is currently on call for an on-call schedule |
connection | A notification connection (Slack channel, webhook, etc.) |
all_admins | Every administrator in the organization |
Repeats
A policy can repeat its whole sequence a configurable number of times, with a delay between repeats — useful for "keep paging until someone acknowledges" behavior.
How It Fits Together
Acknowledging the incident stops the escalation. See Incidents for acknowledgment, snooze, and resolution.