Incident alerts, without the assembly required
Cloudflare’s infrastructure now handles an average of more than 46 million HTTP requests per second, and its customer base includes 31% of the Fortune 1000. When something goes wrong on that scale, the ripple effects are felt across a large swath of the internet. The company has long published a public status page and offered an RSS feed and a Status API for those who want programmatic access to incident data, but those channels put the monitoring burden on the customer. The new Cloudflare Incident Alerts shift that model from pull to push.
The feature is available to any Cloudflare account, including free ones, and delivers notifications via email, webhook, or PagerDuty. Configuration happens entirely in the dashboard, with no custom infrastructure required.
Filling the proactive notification gap
During an incident, the status page is updated with reports, updates, and resolutions, and it also tracks scheduled maintenance. But relying on customers to periodically check that page—or build their own tooling around the Status API or RSS feed—creates a lag between when an incident begins and when affected teams learn about it. The Status API and RSS feed also lack simple filtering options, so subscribers see noise from scheduled maintenance and incidents that don't affect them.
Incident Alerts address that gap with granular, customizable push notifications. Instead of monitoring Cloudflare's status yourself, you choose which products and severity levels warrant a notification, and the alert comes to you.
Filtering by service and severity
Notifications can be scoped along two dimensions: specific Cloudflare products or product areas, and impact level (critical, major, and/or minor). These filters are combinable, so configurations can be tuned to your particular exposure. For example:
- Email for all critical incidents
- Webhook for critical and major incidents affecting Pages
- PagerDuty for all incidents affecting Stream
A Cloudflare for SaaS customer might only want to know about delays in custom hostname activation, not about latency on Stream. Similarly, a team that only needs to react to severe outages can opt out of minor incident notifications entirely. If a filter field is left blank, it defaults to all impact levels or all components.
The dashboard exposes more than fifty notification types beyond incident alerts, each of which supports independent delivery method and filtering configuration.
Setting up an alert
To create an incident alert, navigate to the Cloudflare dashboard's Notifications section, select “Add,” then choose “Incident Alerts.” From there:
- Enter a notification name and description.
- Select the impact level(s) and component(s) you care about.
- Choose your delivery channels: PagerDuty, a webhook, and/or an email recipient.
- Save the configuration.
- Use the “Test” button on the notification's row to verify it fires correctly.
The status page, Status API, and RSS feed remain available for those who prefer those channels. Full documentation for the notification system is available in the Cloudflare developers portal.
