CaryvaneHelp
/

Notifications and webhooks

Caryvane sends the alerts itself; there is no mail server to configure. Choose what to be told about, where, and which jobs and machines are allowed to stay quiet.

Set up #

  1. Notifications in the console (per customer). Add one or more email addresses.
  2. Choose events: Job completed successfully (usually off — it is noisy), Job failed, Agent goes offline. A storage target whose credentials stop working also raises an alert.
  3. Send test proves the route before you rely on it.

Webhooks #

Enter a URL and an HMAC signing secret. Each event is POSTed as JSON — the event type, customer, job or agent, status, message and time — with an HMAC-SHA256 of the body in a signature header computed with the secret, so the receiver can verify the call came from Caryvane. PSA and ticketing systems that accept inbound webhooks (or a small relay) can open tickets on failures this way. Send test posts a sample event.

Mutes #

A job that is expected to warn every night (a folder with EFS files nobody will decrypt) and a laptop that is off every weekend would otherwise drown the real alerts. Mute notifications for this job on the job's Options step, and Notifications… on the agent's menu, override the customer-wide settings for that one thing. Overrides are listed on the Notifications page so they are not forgotten.

What "offline" means #

An agent is online while it has checked in within 90 seconds. The offline alert fires when a machine that was online stops checking in and stays away — not on every laptop lid-close. Machines that are legitimately intermittent should be muted, not tolerated.

For MSPs #

Notifications are per customer: an MSP receives alerts for each customer whose Notifications page lists the MSP's address. For a single feed across all customers, point every customer's webhook at one endpoint and route on the customer field in the payload.