Incident Manager, Pro
Declare an incident, name its commander, post status page updates with a preview, then draft a cited postmortem from the timeline with AI.
An outage run from one block, from declaring it through status page updates to a postmortem drafted by AI.
Open Incident
An open incident counting up in its current stage, with how long each stage took, a one-line update with a status page preview, and a postmortem whose sentences cite the timeline.
Elevated API error rates
Declared 14:02 UTC by Ana Silva. Ben Okafor is commander.
0:10:00
Open for 48 min. Webhooks is running slowly.
- Investigating, 16 min
- Identified, 22 min
- Monitoring, Now
- APIOperational
- WebhooksDegraded performance
- 3 other services operational
Post an update
Timeline
- 1
InvestigatingAna Silva
We’re investigating increased errors on API requests. Some requests return a 502.
- API: Partial outage
- Webhooks: Degraded performance
- 2
CommanderAna Silva
Ben Okafor is incident commander.
- 3
NoteBen Okafor
Error rate on api-gateway is 18%. It started at 13:58, two minutes after load balancer config v42 shipped to us-east-1.
- 4
IdentifiedBen Okafor
We traced the errors to a load balancer change in us-east-1 and are rolling it back.
- API: Partial outage
- Webhooks: Degraded performance
- 5
NoteDev Patel
Rolled back to config v41. api-gateway errors are under 0.5%. 41,000 webhook deliveries are queued and draining.
- 6
MonitoringBen Okafor
The fix is out and API errors are back to normal. Webhook deliveries are delayed while the queue drains.
- Webhooks: Degraded performance
Declare Incident
The form to declare an incident when none is open.
Declare an incident
Name it, pick who leads it and mark what’s affected. The first update goes on your status page.