Alerts & Notifications
Deliver actionable incident alerts to the right people through the channels they already use — Telegram, Slack, email, and SMS.
Explore Channels Escalation PoliciesFour channels, one unified alerting engine
UptimeGuard ships with native integrations for Telegram, Slack, SMTP email, and SMS via Twilio. Each channel supports per-check routing, so your payment gateway downtime can page your on-call engineer while a blog 404 goes to the content team.
Telegram
Send alerts to any Telegram group, channel, or individual chat. Configure a bot token once, then reference chat IDs per-check. Supports rich Markdown formatting, custom alert payloads, and inline reply-to-acknowledge actions. Average delivery latency under 1.2 seconds.
Slack
Post incident notifications to Slack channels via incoming webhooks or the UptimeGuard app. Each message includes the check name, response time, HTTP status, and a deep link to the incident timeline. Threaded follow-ups keep the channel clean while preserving full context.
Deliver alerts through your own SMTP server or UptimeGuard's relay. Define recipient lists per check, set up digest emails for non-critical monitors, and include HTML-formatted reports with uptime charts and last-known-good response bodies.
SMS
When downtime hits critical services, SMS cuts through. Connect a Twilio account or use UptimeGuard's built-in carrier network. SMS alerts include the incident severity, affected endpoint, and a short URL to the dashboard. Rate-limited to prevent alert storms.
Automatic escalation when the first responder does not act
Define multi-level escalation chains that promote an alert if no one acknowledges it within a set window. Typical setups start with a Slack ping to the on-call rotation, escalate to Telegram + SMS after five minutes, and finally notify the engineering manager after fifteen.
Level 1 — Immediate
Alerts fire to the primary channel assigned to each check. For example, a failed health check on api.example.com sends a message to #ops-incidents in Slack and a Telegram notification to the on-call engineer's private chat. Acknowledgment resets the timer.
Level 2 — Escalated
If the Level 1 alert remains unacknowledged after the configured delay (default: 5 minutes), UptimeGuard routes the same incident to the escalation contacts. This typically adds SMS to the delivery set and notifies a secondary Slack channel such as #engineering-all.
Level 3 — Management
After a further delay (default: 15 minutes), the incident escalates to management-level contacts. These recipients receive a consolidated summary via email and SMS, including incident duration, affected checks, and a direct link to the live incident page for rapid triage.
Escalation policies are attached at the check level or inherited from the parent project. You can create separate policies for production, staging, and third-party dependency monitors so that a downed payment provider triggers a different chain than an internal dev server.
Suppress alerts during planned maintenance
Schedule maintenance windows to pause alerting for specific checks or entire projects. When a window is active, UptimeGuard continues monitoring and recording data — it simply suppresses all notifications and marks the period as "planned downtime" on the timeline.
One-time windows
Define a start time and duration for a single maintenance event. Ideal for database migrations, SSL certificate renewals, or firewall rule changes. The window auto-closes once the end time is reached, and alerting resumes immediately. All checks under the window are suppressed.
Recurring windows
Set repeating schedules for predictable maintenance — weekly patching on Sunday 02:00–04:00 UTC, monthly backup windows on the first Saturday, or daily cache rebuilds at 03:00 UTC. Recurring windows support timezone selection so your team sees the correct local time in the dashboard.
Selective suppression
Not every check needs silencing during maintenance. Attach a maintenance window to a subset of checks — for example, suppress alerts on the database health endpoint while keeping the CDN and API gateway monitors fully active. This prevents false positives without losing visibility on unrelated services.
Maintenance windows appear as shaded regions on every check's uptime chart. Historical reports automatically exclude planned downtime from availability calculations, so your SLA metrics reflect only unexpected outages. Audit logs record who created each window and when it was active.
Alerts & Notifications
Deliver actionable incident alerts to the right people through the channels they already use — Telegram, Slack, email, and SMS.
Explore Channels Escalation PoliciesFour channels, one unified alerting engine
UptimeGuard ships with native integrations for Telegram, Slack, SMTP email, and SMS via Twilio. Each channel supports per-check routing, so your payment gateway downtime can page your on-call engineer while a blog 404 goes to the content team.
Telegram
Send alerts to any Telegram group, channel, or individual chat. Configure a bot token once, then reference chat IDs per-check. Supports rich Markdown formatting, custom alert payloads, and inline reply-to-acknowledge actions. Average delivery latency under 1.2 seconds.
Slack
Post incident notifications to Slack channels via incoming webhooks or the UptimeGuard app. Each message includes the check name, response time, HTTP status, and a deep link to the incident timeline. Threaded follow-ups keep the channel clean while preserving full context.
Deliver alerts through your own SMTP server or UptimeGuard's relay. Define recipient lists per check, set up digest emails for non-critical monitors, and include HTML-formatted reports with uptime charts and last-known-good response bodies.
SMS
When downtime hits critical services, SMS cuts through. Connect a Twilio account or use UptimeGuard's built-in carrier network. SMS alerts include the incident severity, affected endpoint, and a short URL to the dashboard. Rate-limited to prevent alert storms.
Automatic escalation when the first responder does not act
Define multi-level escalation chains that promote an alert if no one acknowledges it within a set window. Typical setups start with a Slack ping to the on-call rotation, escalate to Telegram plus SMS after five minutes, and finally notify the engineering manager after fifteen.
Level 1 — Immediate
Alerts fire to the primary channel assigned to each check. For example, a failed health check on api.example.com sends a message to #ops-incidents in Slack and a Telegram notification to the on-call engineer's private chat. Acknowledgment resets the timer.
Level 2 — Escalated
If the Level 1 alert remains unacknowledged after the configured delay (default: 5 minutes), UptimeGuard routes the same incident to the escalation contacts. This typically adds SMS to the delivery set and notifies a secondary Slack channel such as #engineering-all.
Level 3 — Management
After a further delay (default: 15 minutes), the incident escalates to management-level contacts. These recipients receive a consolidated summary via email and SMS, including incident duration, affected checks, and a direct link to the live incident page for rapid triage.
Escalation policies are attached at the check level or inherited from the parent project. You can create separate policies for production, staging, and third-party dependency monitors so that a downed payment provider triggers a different chain than an internal dev server.
Suppress alerts during planned maintenance
Schedule maintenance windows to pause alerting for specific checks or entire projects. When a window is active, UptimeGuard continues monitoring and recording data — it simply suppresses all notifications and marks the period as "planned downtime" on the timeline.
One-time windows
Define a start time and duration for a single maintenance event. Ideal for database migrations, SSL certificate renewals, or firewall rule changes. The window auto-closes once the end time is reached, and alerting resumes immediately. All checks under the window are suppressed.
Recurring windows
Set repeating schedules for predictable maintenance — weekly patching on Sunday 02:00–04:00 UTC, monthly backup windows on the first Saturday, or daily cache rebuilds at 03:00 UTC. Recurring windows support timezone selection so your team sees the correct local time in the dashboard.
Selective suppression
Not every check needs silencing during maintenance. Attach a maintenance window to a subset of checks — for example, suppress alerts on the database health endpoint while keeping the CDN and API gateway monitors fully active. This prevents false positives without losing visibility on unrelated services.
Maintenance windows appear as shaded regions on every check's uptime chart. Historical reports automatically exclude planned downtime from availability calculations, so your SLA metrics reflect only unexpected outages. Audit logs record who created each window and when it was active.