Coalesce provides several mechanisms for viewing the results of a refresh to leverage in troubleshooting, as well as initiating a re-run upon failure.
Viewing the Results of a Refresh
The results of a refresh are available directly within your Workspace in the Coalesce app (UI) for an object run within a Workspace, and via the Activity Feed in the UI, the Command Line Interface (CLI) log, and/or the REST API via the results endpoint for the refresh of a Job in a deployed Environment.
Activity Feed
The Activity Feed in the Coalesce app (UI) provides details of both deploys and refreshes within your organization and is available via the right panel of the Deploy Interface.
You are able to apply filters to the results to limit by Project, Environment, type (deploy and/or refresh), and status.
The Activity Feed provides full details for each Refresh run in a visual interface. These details include:
- Connection details
- Parameters used
- Details for each Node + steps for a Node
- Snowflake query ID
- Error message(s), if any
Command Line Interface (CLI)
The same details available in the Activity Feed about a refresh are logged by the CLI when the CLI is leveraged for refresh and can be actioned upon depending upon your requirements and orchestration method. See the CLI Refresh section of the Coalesce documentation for more information.
REST API
The same details available in the Activity Feed about a refresh are available via the API using the results endpoint. See the List Run Results section of the Coalesce documentation for more information.
Error Troubleshooting
In the case of any errors, Coalesce surfaces detailed error messages at the level of the step for the Node that failed; the developer can then use these error messages and the other associated information, such as the Snowflake query ID and SQL, to identify and resolve the source of the error.
Running from Point of Failure
In the case of refresh failures for deployed Environments, Coalesce provides a mechanism for rerunning a refresh from the point of failure. If the failure was due to an error that requires manual resolution, you will need to complete this resolution prior to re-initiating the refresh to ensure it completes successfully on rerun.
- In the CLI, this is completed via the coa rerun command. See the CLI Failed Refreshes / Jobs section of the Coalesce documentation for more information.
- In the API, this is completed via the rerun endpoint. See the Trigger a Job to Rerun via the API section of the Coalesce documentation for more information.