---
title: "Incidents and maintenance"
description: "Nextvision Status — Incidents and maintenance"
---

> Documentation Index
> Fetch the complete documentation index at: https://status.nextvisions.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Incidents and maintenance

## Automatic detection

A confirmed monitor failure changes its linked component and creates an unpublished incident draft. Operators receive an internal email/webhook alert. The public page receives no unverified narrative.

## Publish an incident

Open Incidents, review the draft, select affected components and language, and write an update. Check Published and save. Updates progress through Investigating, Identified, Monitoring and Resolved. Published incidents cannot be unpublished. Use a correction update to preserve the timeline.

Write a verified root cause only when evidence supports it. Add remediation and postmortem findings after investigation. Internal notes and attachments are not published through the public API. The public narrative is not automatically translated.

## Maintenance

Create a maintenance window with affected components and an IANA timezone. Date inputs use the operator browser's local time; the UI sends absolute UTC timestamps. The timezone field is display context, not a second conversion of those timestamps. Maximum duration is 31 days. Edit the body/window to record progress or end a maintenance early.

Within the window, components show Maintenance and measurements continue. When the window ends, the current monitor state becomes visible again.

## Notifications

Subscribers confirm by email and can unsubscribe using the link in each update. Confirmation links expire after seven days. Operators should keep credentials and infrastructure addresses out of public update text.

Source: https://status.nextvisions.ai/docs/en/incidents/index.mdx