Viewing Task History in Snowsight¶
Tasks let you schedule the execution of SQL code. You can create a Directed Acyclic Graph (DAG) of tasks to specify how to execute SQL code specified in a root task and a set of dependent tasks. A task is associated with a specific database and schema. For more details about tasks, refer to Introduction to Tasks.
Using Snowsight, you can review the execution history for tasks in several different ways:
View the execution history of all tasks run in your account, for example to identify critical tasks that failed to run, long-running tasks, or tasks increasing costs.
View the execution history for a specific task, to gather more information about the task.
Required Privileges for Viewing Task History¶
To view task history in Snowsight, you need the same privileges required to run the TASK_HISTORY and COMPLETE_TASK_GRAPHS table functions.
You must use one of the following:
The ACCOUNTADMIN role.
A role with the OWNERSHIP privilege on the task.
A role with the MONITOR EXECUTION global privilege, and USAGE privilege on the database and schema that store the task.
Viewing Individual Task Graphs¶
Review a task graph to see a root task and its dependent tasks in the format of a DAG. When you review a task graph, you can do the following:
View task information, including status by root task and selected dependent task.
Examine the task graph, including the root and dependent tasks in the graph. Refer to DAG of Tasks.
Select a task on the graph to view additional details, such as predecessor tasks, the warehouse used to run the task, and the role that owns the task.
To view a task graph for a specific database schema, do the following:
Sign into Snowsight.
In the navigation menu, select Data » Databases.
Use the object explorer to locate the database and schema that contain the tasks that you want to view.
For the selected schema, select Tasks.
Select a specific task.
The task details appear, with additional Graph, and Run History tabs.
Select the Graph tab to view the task graph.
The task graph appears, displaying a hierarchy of tasks.
Select a task to view details in the context of the graph.
Note
Task history data is only available if the task has been executed in the last 7 days.
Review the Run History for a Task¶
Task run history includes details about each execution of a given task. You can view the scheduled time, the actual start time, duration of a task and other information.
For each instance, you can view the following:
Scheduled start time: When is the task next scheduled to run.
Status: Status of the most recent run.
Duration: Amount of time, in seconds, for the most recent run.
Query: View the task query profile or task statements in a worksheet.
To view the run history:
In the Snowsight navigation menu, click Data » Databases.
In the right pane, using the object explorer, navigate to a database and schema.
For the selected schema, select and expand Tasks.
Select a task. Task information is displayed, including Task Details, Graph, and Run History sub-tabs.
Select the Run History tab.
Note
Task history data is only available if the task has been executed in the last 7 days.
Review Account-Level Task History¶
Review the account-level history for task runs to identify failing tasks, long-running tasks, and other monitoring and debugging cases for an entire account, rather than for one specific task.
To view account-level history for tasks, do the following:
Sign into Snowsight.
In the navigation menu, select Activity » Task History.
To view individual task runs, select View » Task Runs from the filters.
After you select the history of task runs, you can filter the page to display relevant information.
Select the Date Range filter to show task history from the last day through the last 12 months, or a custom range.
Select the Task status filter to display task history for one or more status, such as Succeeded, Failed, Cancelled, or Skipped.
Filter on the name of the task to see patterns in status or duration over time for specific tasks.
Filter on the name of the database or schema that contain the tasks.
For example, to identify long-running tasks, select the Status filter to show only successful tasks, and sort the Duration field in descending order. For advanced debugging, you can open the filtered and sorted table in worksheets using the Open in worksheets button. You could then modify the SQL statement with LIMIT / FETCH and GROUP BY arguments to identify the databases and schemas with the top 25 most long-running tasks.
You can also select a specific task to drill down for more details.