Alerting Overview

Hoppla alerts you when something goes wrong with your monitored services. Alerting is built on three concepts: notification channels (where alerts are sent), notification rules (when alerts are sent), and incidents (what triggered the alert).

How Alerting Works

When a check detects a problem, the following happens:

1. Check fails -- A monitoring check returns an unexpected result.
2. Verification -- For uptime checks, the failure is verified from a second probe location before proceeding. Single-location failures do not trigger alerts.
3. Alert evaluated -- Hoppla evaluates the failure against configured thresholds and rules.
4. Incident created -- If the alert criteria are met, an incident is created.
5. Rules matched -- Notification rules are checked to determine which channels should be notified.
6. Notifications sent -- Alerts are dispatched to all matching channels (email, Slack, webhook, etc.).
7. Escalation (optional) -- If a rule has an escalation policy attached and the incident isn't acknowledged in time, Hoppla escalates it through additional steps -- optionally paging whoever is on call. Acknowledging the incident stops further escalation.

When the check passes again, the incident is automatically resolved and a resolution notification is sent through the same channels.

Event Types

The following event types can trigger notifications:

Event	Description
Incident Created	A monitor detected a failure (downtime, expired certificate, etc.)
Incident Resolved	A previously failing monitor recovered
Monitor Flapping	A monitor is rapidly switching between up and down
Flapping Resolved	A previously flapping monitor has stabilized
Incident Storm Detected	Multiple monitors failing simultaneously
Storm Resolved	An incident storm has cleared
SSL Expiring	SSL certificate expiring soon
Domain Expiring	Domain registration expiring soon
Degraded Performance	Lighthouse scores below configured threshold
Application Health	Health endpoint reporting issues
Sitemap Issues	XML sitemap validation problems
Security Headers	Missing or misconfigured security headers
Redirect Chain	Redirect issues detected (loops, excessive hops)
Robots.txt	Robots.txt problems detected
DNSBL Listing	Server IP found on a DNS blacklist
Broken Links	Broken links detected during a crawl
Mixed Content	Insecure (HTTP) content detected on an HTTPS page
Server Resources	Agent metrics exceed configured thresholds
Warning Cleared	A previous warning has resolved
DNS Records Changed	DNS record changes detected, including changes to the zone's authoritative nameservers (an NS-hijack signal). The DNS change payload includes the zone, nameservers, and the per-record changes.
Certificate Renewed	SSL certificate was renewed

Quick Setup

To start receiving alerts:

1. Create a notification channel -- Configure where alerts go (email, Slack, webhook, etc.). See Notification Channels.
2. Create a notification rule -- Define which events trigger notifications on which channels. See Notification Rules.
3. Add monitors -- Your monitors will now generate incidents and notifications automatically.

By default, new teams have an email channel pre-configured for the team owner's email address.

Going further

• Escalation policies -- Make sure critical incidents that nobody acknowledges keep climbing the chain until someone responds.
• On-call schedules -- Rotate who gets paged by day and time, so the right person is alerted instead of the whole team.