> ## Documentation Index
> Fetch the complete documentation index at: https://fal.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.
# Slack Notifications
> Receive real-time alerts in Slack when your apps fail to start.
Connect Slack to your fal account to receive notifications when your applications encounter startup issues. Get alerted immediately when a runner fails, so you can investigate before users are affected.
## Setting Up Slack
Navigate to [**Dashboard > Notifications > Settings**](https://fal.ai/dashboard/notifications/settings)
Click **Add to Slack** to begin the OAuth flow. This installs the fal bot into your Slack workspace.
Select the Slack channel where you want to receive notifications. The fal bot will automatically join this channel.
Any authenticated team member can connect Slack to receive notifications.
## What You'll Be Notified About
### App Startup Timeout
Triggered when a runner's `setup()` method doesn't complete within the configured `startup_timeout`. The notification includes:
* App name and runner ID
* The timeout duration that was exceeded
* Link to logs for investigation
* Link to the startup timeout documentation
This typically indicates your model loading or initialization is taking too long. Consider optimizing your `setup()` method or increasing the `startup_timeout` value.
### App Startup Failure
Triggered when a runner's `setup()` method throws an error. The notification includes:
* App name and runner ID
* Link to logs to see the full error traceback
Common causes include missing dependencies, incorrect model paths, or out-of-memory errors during model loading.
### Queue Size Threshold
Triggered when the number of requests waiting in your app's queue exceeds a threshold over a time window. This usually means your app isn't scaling fast enough to meet demand, or callers are sending more traffic than your `max_concurrency` can handle.
### HTTP 5xx Threshold
Triggered when the number of HTTP 5xx responses from your app exceeds a threshold over a time window. A spike in server errors often points to a crashing endpoint, a bad deployment, or resource exhaustion.
## Per-App Notification Settings
You can configure alert thresholds for each app from the dashboard. These control when queue size and error rate notifications fire.
| Setting | Default | Description |
| ------------------------------- | ------- | ------------------------------------------------- |
| `queue_size_threshold` | 100 | Number of queued requests that triggers an alert |
| `queue_size_threshold_duration` | 1h | Lookback window for queue size (e.g., `5m`, `1h`) |
| `http_5xx_threshold` | 100 | Number of 5xx responses that triggers an alert |
| `http_5xx_threshold_duration` | 1h | Lookback window for 5xx count (e.g., `5m`, `1h`) |
Set a threshold to `null` to disable that notification type. Lower durations (e.g., `5m`) make alerts more sensitive to short spikes; longer durations (e.g., `1h`) smooth out transient errors.
## Managing Your Connection
From the [Notification Settings](https://fal.ai/dashboard/notifications/settings) page you can:
* **View connection status** -- See if Slack is connected, and which workspace and channel are configured
* **Change channel** -- Switch notifications to a different Slack channel
* **Reconnect** -- If the bot token expires or the bot is removed from your workspace, reconnect with one click
## Troubleshooting
**Not receiving notifications?**
* Verify the fal bot is still in your selected channel (it may have been removed)
* Check that your Slack connection hasn't expired in the notification settings
* Ensure the app experiencing issues belongs to your team account
**Bot was removed from workspace?**
* Go to notification settings and click **Reconnect** to re-authorize the fal bot
View real-time metrics and logs in the fal dashboard