Slack Notifications

Using Slack’s Incoming Webhook, you can send monitor incident notifications to a Slack channel.

1. Create a Slack App

   Go to Slack API: Applications, select “Create New App”, then choose “From scratch”. Specify an App Name and the workspace where you want to send notifications.

2. Enable Incoming Webhooks

   On the settings page of the app you created, open “Incoming Webhooks” and turn the toggle ON.

3. Generate a Webhook URL

   Click “Add New Webhook to Workspace”, select the channel you want to send notifications to, and click “Allow”.

4. Copy the Webhook URL

   Copy the URL displayed in the format https://hooks.slack.com/services/T.../B.../xxx.

1. In the dashboard, open Settings > Integrations, then select Configure for Slack
2. Enter a recognizable Name
3. Paste the Incoming Webhook URL into Slack Webhook URL
4. Select Add

Slack notifications are sent as text-format messages. Emojis are sent in Slack alias format (e.g., :red_circle:) and rendered on Slack.

:red_circle: *{Monitor Name}* is DOWN
Incident: {Incident ID}
URL: {Monitored URL}

:large_green_circle: *{Monitor Name}* is back UP
Incident: {Incident ID}

Maintenance start and end events include the affected monitors; the start message also includes the scheduled end time.

• The URL is rejected during setup: Manako accepts only HTTPS URLs beginning with https://hooks.slack.com/. Webhook URLs on other hosts, including GovSlack, are not supported.
• The message goes to the wrong destination or does not arrive: An Incoming Webhook is bound to the conversation selected when it was issued. Confirm that the Slack App and Webhook still have access to that conversation.
• A test arrives but normal notifications do not: When a monitor has channel links, only active linked channels are used. If links exist but all are disabled, Manako does not fall back to team-wide channels.
• Notifications stop during rapid state changes: After more than three state changes for the same monitor within 10 minutes, the flap guard suppresses subsequent notifications.
• Notifications appear duplicated: Creation and recovery are separate events. A sent retry for the same incident, channel, and event type is skipped; failed or pending sends are retried.

Test sends an actual Slack message and has a 60-second cooldown after success. A failure on one channel does not stop delivery to the others.

The flap guard and incident-level sent check apply to incident notifications. Maintenance notifications bypass both controls and go to active, verified team channels.