Skip to main content

Update Incident Details

As you investigate and work toward resolution, keep the incident record current.

Accurate details help stakeholders stay informed, trigger the right automated workflows, and create a reliable record for post-incident review.

Edit Incident Fields​

  • Use the pencil icon next to any field to edit it inline (severity, assignee, affected services, etc.).
  • Click Edit on the Incident Summary to update the description with new findings or context.
  • Click Save after making changes.

Update fields whenever the situation changes — for example, if you discover additional affected services or need to reassign the incident to another responder.

Update Incident Status​

Keep the status current as you work through the incident. Status changes are visible to all stakeholders and may trigger automated workflows.

StatusMeaning
InvestigatingYou're actively looking into the issue.
IdentifiedRoot cause or a contributing factor has been found.
MonitoringA fix has been applied; you're watching for stability.
ResolvedThe incident is closed.

Change the status from the incident details page by clicking the status field and selecting the new value.

Incident status dropdown

Add Key Events​

Key events mark important milestones: root cause identified, mitigation applied, service restored, a decision to escalate, an external communication sent, etc.

  1. Click Add Key Event on the incident details page.
  2. Type a description of what happened.
  3. Click the check mark to save.
  4. Click Save from the top right.

Key events appear in the incident timeline and are used by the AI Scribe Agent when generating post-incident summaries. Adding them as they happen — rather than reconstructing after the fact — produces a more accurate record.

Send Status Updates​

Keep stakeholders informed without interrupting the incident response by sending structured status updates via email. Status updates are delivered to all users and teams subscribed to the incident's impacted services.

How to Send a Status Update​

  1. Open the Incident Details page.
  2. Click Status Update (typically in the top action bar).
  3. Review the pre-populated email template:
    • Subject: Default format is "AI SRE Status update for incident [incident-ID]"
    • Message body: Includes incident ID, title, summary, impacted services, and current status
  4. Edit the subject and message as needed:
    • Add business context (customer impact, revenue impact, etc.)
    • Include mitigation actions taken
    • Set expectations for next steps and timeline
  5. Review the recipient preview:
    • The system shows all resolved recipients and the total email count
    • Recipients are automatically gathered from all impacted services' subscriber lists
    • Duplicates are removed if a stakeholder is subscribed to multiple impacted services
  6. Click Send.

When to Send Status Updates​

Send updates at significant milestones during the incident:

  • Status change: When moving from Investigating → Identified → Monitoring → Resolved
  • Mitigation applied: After deploying a fix or implementing a workaround
  • Scope change: When additional services are impacted or the severity changes
  • Customer impact: When customer-facing impact begins or ends
  • Extended incidents: For long-running incidents, send periodic updates even if status has not changed (e.g., every 2 hours)

Email Delivery​

  • Delivery channel: Email only (from aisre-noreply@harness.io)
  • Format: Branded HTML email with header and footer images
  • Timeline record: A "Status Update Sent" event is added to the incident timeline when delivery succeeds
Reduce Interruptions

Status updates allow stakeholders to stay informed without joining the incident war room or interrupting responders with "what's the status?" messages. Send updates proactively at key milestones.

Best Practices​

  • Update status as soon as the phase changes — don't wait until resolution to batch-update. Real-time status drives stakeholder confidence and automated workflows.
  • Be specific in key events — "Identified root cause: connection pool exhaustion on db-primary-01" is far more useful than "Found the issue."
  • Reassign when appropriate — If you've identified that the incident belongs to another team or specialist, update the assignee and notify them.