An alert rule decides what's worth a notification and where it lands. The moment a monitor changes state, SolidOps delivers, and tells you again when it recovers. Live today.
An alert rule decides what counts as worth a notification and where it lands. From there a state change runs the loop on its own, all the way back to healthy.
One rule, read top to bottom: what makes it fire, where the notification goes, and that a recovery follows on the same path.
A monitor crosses into down or degraded. SolidOps marks the change the instant a check sees it, and the matching rule fires.
The notification goes out to the channels on the rule. Today that means email, a generic webhook, or both at once.
When the monitor reads healthy again, a recovery notification follows on the same channels, so you know it's back, not only that it broke.
Two delivery channels are live today: email and a generic webhook, on their own or together. More channels are being built and are clearly marked as coming.
A plain, readable email the moment a monitor changes state, and again when it recovers. The first channel anyone can use, with nothing to wire up.
A generic webhook posts the event to any URL you control, so you can route alerts into your own tools, scripts, or an internal handler.
{"event": "monitor.down","monitor": "billing.acme.io","status": 503,"state": "down","incident": "inc_4c1a","at": "2026-06-22T14:02:08Z"}
Alerts posted into a channel your team already watches.
Not built yet
Delivery to a Telegram chat or group, for teams that live there.
Not built yet
A push notification straight to your phone, away from the inbox.
Not built yet
These rows are placeholders for what we're building. They don't deliver anything yet. Email and webhook are the only channels you can pick today.
An alert that no one acknowledges would escalate to other channels after a delay you set, so a missed first notification doesn't become a missed incident. This is on the roadmap and is not built yet.
Set a delay and a fallback channel. If the alert isn't acknowledged in that window, it steps up to the next rung on its own, past the first channel, out to the next.
The first notification goes out on the rule's primary channel.
No one has acknowledged it, so the alert would step up to a second channel.
Still open. It would reach further out, to whoever is next on the ladder.
Illustration of a planned feature. None of these steps run today. Alerts deliver once, to the channels on the rule.
Alerts fire on every state change and deliver to email or a webhook today, with a recovery notification once it's healthy again. More channels and escalation are on the way, in the open.