Create and manage sensors

Create a sensor#

Click New monitor — in the sidebar, on the overview, or on the Resources page — and choose how to start:

• Start from a template — pick a common scenario (Website availability, Nightly backup, PostgreSQL availability…): the form below opens pre-filled, you complete what is specific to you. See Templates.
• Configure manually — the wizard asks four questions, described below, and gives access to every type and option.

1. What to monitor#

Pick a goal: Website, API, Database, Scheduled job, Server, Network, SSL certificate, Domain name, Message queue or Files. You do not need to know the sensor types: the goal leads to them.

Server is not created here: server sensors come with each private agent, so this card simply leads to the agents page.

2. Strategy#

Choose the check that matches what matters to you. For example:

Goals with a single check skip this step. A check that is not included in your plan shows an Upgrade button.

3. Configuration#

The configuration step has the same sections for every type:

• General — Name, Service and Environment (both pre-filled with your defaults) and optional Tags (press Enter or a comma to add one).
• Schedule and severity — see below.
• The type-specific settings — the request, the host, the query… Each type's page details them.
• Agent — the Onagre public agent, selected by default, or one of your private agents. Only Advanced agents are listed; each one shows whether it is Online. Types that need a private agent hide the public one.
• Alerting — the integrations to notify. If you select none, Onagre asks you to confirm: the sensor will open incidents but warn nobody.

4. Review#

Check the summary — each part has an Edit button — then click Create monitor. You land on the sensor's page, where the first result appears after the first check.

Schedule and severity#

Check schedule

• Interval — every 1, 2, 5, 10, 15 or 30 minutes, or every 1, 2, 6 or 12 hours. The Free plan starts at 5 minutes; the default is 5 minutes.
• Daily — once a day, at a given time.
• Weekly — once a week, on a given day and time.

Times are in your organization's time zone (set in your profile). Certificate checks always run once a day.

Severity — Critical, Major, Minor or Info. It is copied onto every incident the sensor opens, so you can sort and filter incidents by importance. It does not change who is notified.

Timeout — for network and database checks, the time allowed before the check fails, in milliseconds. It defaults to your plan's maximum: 5 seconds on Free, 15 seconds on paid plans.

Hosts and addresses#

Fields called Host accept a hostname (db01), a fully qualified domain name (api.corvane.example), an IPv4 address (10.20.1.21) or an IPv6 address (2001:db8::1). Do not include a scheme (https://) or a path: those belong in URL fields.

Remember where the check runs: the public agent only reaches public addresses, so 10.20.1.21 or db01 need a private agent inside the same network.

Reuse values across environments#

Fields such as URLs, hosts, headers, SQL queries and Redis keys accept variables written ${NAME}, taken from the sensor's environment: https://${API_HOST}/health checks a different host in Staging and in Production. Database, key-value and queue sensors use secrets for their connection strings.

Using a variable or a secret in a sensor requires the Manager role. See Variables and Secrets.

Manage existing sensors#

From a sensor's page:

• Edit reopens the wizard on the configuration step; Save changes applies it from the next check.
• Pause suspends the checks (no results, no incidents) until you click Start.
• More actions → Duplicate creates a copy and opens it for editing; Delete removes the sensor and its history. An open incident is closed as resource deleted.

From the Resources list, select several sensors with their check boxes to Start, Pause or Delete them at once, or to set their integrations with Alertings (this replaces their current alerting). Click a sensor's state to pause or start it.

To find sensors, filter by type with the chips at the top, by state with All / Up / Down / Paused / Created, or search by name. The search box also understands tag:checkout and state:down.

Server and agent heartbeat sensors are managed with their agent: they cannot be paused, duplicated or deleted on their own.

Who can do what#

Viewers can see sensors and their results. Creating, editing, pausing and deleting sensors requires the User role; using variables or secrets requires Manager. See Team and roles.

Limits#

Each plan includes a number of sensors — 10 on Free, then 50 to 1,000. Server sensors do not count. When the limit is reached, creation is refused until you upgrade. See Plans.