Snowflake kit
The fastest way to monitor Snowflake. One guided install connects your warehouse and Slack, then creates 12 production-ready metrics automatically, no manual dataset or metric setup required.
Where to find it
In the app, go to System → Library and open Snowflake Performance Monitoring. Click Install to start. You may also see this feature referred to as the "Snowflake Monitoring Kit"; it's the same install.
The install flow
There are three steps. The last two are optional and can be skipped.
- 1Connect Snowflake: Create a new Snowflake connection or pick an existing one, then test the connection. Once connected, Lighthouse automatically creates 6 datasets and all 12 metrics for you.
- 2Connect Slack (optional): Authorize the Lighthouse Slack app so metrics can post alerts. You can skip this and connect Slack later.
- 3Choose a notification target (optional): Pick a Slack channel and/or a Slack user to receive alerts. Skip this if you haven't connected Slack, or aren't ready to turn notifications on yet.
Connecting Snowflake
You'll need an account identifier, username, password, database, schema, and warehouse. Role is optional. The connection must pass a live test before you can continue.
For the full field reference and read-only permission setup, see Connect a Warehouse.
What gets created
All 12 metrics are created together; there's no picking or choosing which ones to install.
| Area | Metrics |
|---|---|
| Cost & credits | Total Credits Spike, AI and Other Services Credits Spike, Warehouse Credit Spike |
| Performance | Long-Running Queries, Remote Disk Spill Anomaly, Query Queue Anomaly, Warehouse Utilization |
| Reliability | ETL Failure Rate Spike, Failed Task |
| Storage | Database Storage Spike, Failsafe Storage Spike, Stale Tables |
Changing or removing the kit
You can't select individual metrics after installing. To start over, uninstall the kit (this removes the 12 metrics it created but leaves your Snowflake data source and datasets in place), then install it again.