Alerting, on-call and incident response
Turn signals into alerts that reach the person on call, keep expected noise quiet, and work the incident with the evidence in one place. Rules, schedules, escalation and war rooms share one data store with the telemetry they act on.
- Rule sources
- Metrics, logs and traces
- Notification channels
- Nine channel types
- On-call
- Schedules, overrides, escalation
- Reliability
- SLOs with error budgets
Alert rules
Thresholds on any signal
Threshold rules on metrics, log counts and trace measures, plus rate, ratio and percentile conditions.
Missing data
Absence and no-data rules fire when something that should report goes quiet, such as a heartbeat or a batch job.
Sustain windows
Require a condition to hold for a set time before it fires, so a single spike does not page anyone.
Silences
Mute matching alerts for a maintenance window or a known issue, and let them expire on their own.
DocumentationSeverity and routing
Each rule has a severity, and severity decides which channels and which on-call schedule are notified.
DocumentationThe Alerts page
A live feed with filters and statistics, where alerts are acknowledged and resolved.
Documentation
On-call and SLOs
On-call schedules
Rotations with overrides for holidays and swaps.
DocumentationEscalation policies
If an alert is not acknowledged in time, the next level is notified.
SLOs and error budgets
Define a target for a user-facing measure and track the remaining error budget and how fast it is being spent.
DocumentationFast-burn response
When an SLO budget is burning fast, VerOps opens a war room for it automatically.
Where notifications go
| Channel | Typical use |
|---|---|
| Low-urgency notices and summaries | |
| Slack | Team channels |
| Microsoft Teams | Team channels |
| Discord | Community or small-team channels |
| SMS | Urgent pages to a phone |
| PagerDuty | Hand off to an existing paging setup |
| Opsgenie | Hand off to an existing paging setup |
| Jira | Create an issue to track follow-up work |
| Webhook | Anything else that accepts an HTTP call |
War rooms
A war room is one page for one incident.
Timeline and evidence
A shared timeline, pinned evidence, participants and notes, so people joining late can catch up.
Affected topology
The services and dependencies around the incident, with recent changes overlaid.
DocumentationOpens itself
A war room can open automatically after a set number of consecutive synthetic failures, or when an SLO is burning fast.
Runbooks in the room
Run a workbook from the war room and keep its result with the incident.
Learn more
Read more in the docs
Try it on your own systems
Create a workspace, or tell us about your environment and we will walk you through it.