Skip to main content
Pulse only sees what you send it. This page gets one alert source flowing into Pulse and shows you how to confirm it works before you rely on it.

​
Prerequisites

  • The workspace Admin role. Developers and Viewers can see Resolve → Integrations but can’t add a signal. See Roles.
  • Access to the tool that produces your alerts (for example CloudWatch, Datadog or Grafana), so you can add a webhook or fire a test alert. For a webhook, the tool must be able to reach the webhook URL over HTTPS.
  • For AWS polling: an AWS connection in the workspace. AWS GuardDuty, AWS Config and AWS IAM Access Analyzer must already be turned on in that AWS account, in at least one region. CloudThinker reads them; it doesn’t turn them on.
  • For chat: Slack or Microsoft Teams connected to CloudThinker (Slack, Teams).

​
Pick a signal path

Each platform on Resolve → Integrations can feed Pulse in up to three ways. Pick the one your tool supports. Connect alert sources matches your tool to one of these paths and names the page that sets it up.

​
Connect a source

1

Open the platform

Go to Resolve → Integrations and select your platform. Use the search field if the list is long. For a system with no card of its own, select Custom source.
2

Check the connection

In the Connection section, connect the account or workspace if the platform needs one. Platforms marked Needs connection can’t add signals until this is done.
3

Add a signal

In Signals to Resolve, click Add signal (or Add another signal) and follow the path for your source:
  1. For Custom source, enter a Source name and click Create endpoint first. Copy the Webhook URL and the Secret Key. The Secret Key is shown only once. If you lose it, open the row’s ⋯ menu and click Regenerate credentials. That replaces both the Webhook URL and the Secret Key, and the old pair stops working at once, so update both in your tool.
  2. Paste both into your monitoring tool. The secret authenticates each call in the way your tool supports. The dialog labels it after the Auth Method: Secret Key for Bearer Token, HMAC Secret for HMAC Signature, and API Secret for API Key. Choose authentication covers each method. For CloudWatch, follow the CloudWatch walkthrough, which uses EventBridge.
  3. Send one alert from the tool.
  4. For a vendor platform, the field mapping and severity mapping are preset, but check them against a real payload. Some presets miss values the tool sends: see the Datadog notes in Map the fields. For Custom source, click the payload fields that hold the Incident title (required), Severity, Service, Resource and Description. Paths are dotted, like event.data.title. A leading $. is optional and removed.
Full walkthroughs: CloudWatch and custom source.
4

Send a test alert

Trigger a real or test alert in the source tool, then open Resolve → Pulse and set the lane to All. A polled AWS source shows only after its next poll (see the AWS polling tab). If you resend a test, change its title or wait a minute. Pulse drops an exact copy whose timestamp falls in the same minute as an earlier one. For a webhook, the timestamp is the field Timestamp Path points to, or a top-level timestamp key when no path is set, or the time Pulse received it when there’s neither. So if your tool resends the same event with its original timestamp, waiting doesn’t help: change the title.Success state: the alert appears as a cluster, and the platform shows Ready on Resolve → Integrations. On Slack, Teams and Google Chat, Ready also needs a channel for Post incident updates, or that outcome switched off.
A connected account alone doesn’t prove events arrive. Always confirm one signal in Pulse before you count on a source.

​
Webhook options that change behavior

Open a webhook from its platform’s Signals to Resolve section to edit these. These options combine with the workspace On duty switch in the Resolve → Investigation header. What starts an investigation shows every combination in one table. In every case, a person can still click Escalate on the cluster or Investigate on the incident.

​
Read the readiness badge

Each platform on Resolve → Integrations carries one badge that sums up its whole setup, from the connection to the signal rows and the outcomes. Read readiness lists every badge, what it means and what to do about it. For a signal source, Ready means every connection and source row is set up and active. It doesn’t prove an event arrived, so confirm one signal in Pulse.

​
Troubleshooting

Cause: the source never sent, or sent to the wrong URL. Check: the delivery log in your monitoring tool. Fix: paste the webhook URL again and fire a test alert.
Cause: suppression dropped it, often as a duplicate: same title, severity and resource, with a timestamp in the same minute as an earlier alert. Check: the funnel at the top of Resolve → Pulse. If the raw event count went up but no cluster appeared in the All lane, suppression dropped the event. Fix: send a different alert, or see noise suppression.
Cause: the field mapping points at the wrong path. Fix: edit the webhook, send a fresh payload and pick the field again.
Cause: the AWS connection behind it is unhealthy or lacks permission. Fix: check the AWS connection.

​
Next steps

Work with clusters

Triage your first cluster and learn what each action does.

Webhook integrations

Field mapping, severity mapping and per-vendor setup.

Manage integrations

See every platform’s connection, signals and outcomes.

Resolve settings

Control when CloudThinker investigates on its own.