NAS-C01最も有効な質問と解答で勉強

Consider the following 'manifest.yml' file for a Snowflake Native App:

• A. The application version will default to '0.1.0' and the application can be installed successfully by any consumer account.
• B. The application will fail to install because no 'privileges' are listed in the manifest file. All applications must declare the necessary privileges.
• C. The application version will default to '0.0.1'. If the application includes a view, then the view needs access to the user data and must have a privilege declared to allow users to view data from the application.
• D. The application will fail to install because the version property is missing. Each version must explicitly be specified in the manifest file.
• E. The application version will be automatically assigned by Snowflake during the installation process on the consumer side.

解説: (ShikenPASS メンバーにのみ表示されます)

A data engineering team is developing a Snowflake Native Application that processes sensitive PII dat a. They need to update the application frequently with bug fixes and performance improvements without disrupting user access. Which of the following strategies BEST balances rapid iteration with data security and minimal downtime? Consider that the application uses a secure UDF that needs to be updated in conjunction with other application code.

• A. Implement in-place upgrades to the application package every time a bug fix is released, granting UPDATE privileges on the underlying data tables to the application.
• B. Completely rebuild the application package for each release, pushing a new application to Snowflake Marketplace, forcing users to uninstall and reinstall. This guarantees a clean state but may cause disruptions.
• C. Release patches as separate SQL scripts that consumers must manually apply to their installed application instances using the ACCOUNTADMIN role.
• D. Use versioning, leveraging application roles and grants to control access to specific application versions. Employ automated rollout procedures to progressively expose new versions to consumers using application lifecycle management features.
• E. Use versioning, creating a new version of the application package for each set of changes, and instruct consumers to manually update their application instances to the latest version.

解説: (ShikenPASS メンバーにのみ表示されます)

In a Snowflake Native Application, you are using Streamlit to present data to different user roles. You need to dynamically adjust the Streamlit UI based on the user's privileges. Specifically, if a user has the 'MODIFY' privilege on a particular table, you want to display an 'Edit' button. Otherwise, the button should be hidden. Assuming you are already connected to Snowflake via Streamlit and have the user's role information, which of the following approaches is the most secure and efficient way to determine if the current user's role has the 'MODIFY' privilege on the 'MY TABLE table in the 'MY SCHEMA' schema? (Select two)

• A. Utilize the 'snowflake.permissions' module (if it existed) and call a hypothetical function like 'MY TABLE', 'MY_SCHEMA', user_role)'.
• B. If this hypothetical function return 'TRUE', then the user_role has privilege. Execute the SQL query: 'SELECT COUNT( ) FROM INFORMATION SCHEMA.TABLE PRIVILEGES WHERE GRANTEE = CURRENT ROLE() AND TABLE_SCHEMA = 'MY_SCHEMA' AND TABLE_NAME = 'MY_TABLE' AND PRIVILEGE = 'MODIFY". Check if the count is greater than 0.
• C. Implement a custom stored procedure in Snowflake that encapsulates the privilege check logic. Call this stored procedure from Streamlit, passing the schema, table, and privilege name as parameters. Ensure the stored procedure runs with caller's rights.
• D. Execute the SQL query SELECT HAS PRIVILEGE('MODIFY', 'TABLE', 'MY SCHEMA.MY TABLE')' as the application role. If the result is 'TRUE', then the application role has privilege.
• E. Execute the SQL query SELECT CURRENT ROLE()' to fetch the user's current role, then execute 'SHOW GRANTS TO ROLE and parse the output in Python to check for the 'MODIFY privilege on '

解説: (ShikenPASS メンバーにのみ表示されます)

You're developing a Snowflake Native Application that requires frequent updates to its core logic. You need to implement a deployment strategy that minimizes downtime and allows for easy rollback in case of issues. Which combination of deployment techniques would BEST achieve this goal when deploying updates to your Snowflake Native App? (Choose two)

• A. Implement a blue/green deployment strategy, maintaining two identical environments. Deploy the update to the 'blue' environment, test, and then switch traffic from 'green' to 'blue'. If issues arise, switch back to 'green'.
• B. Implement canary deployments by gradually rolling out the new version to a small subset of consumers, monitoring performance, and then expanding the rollout if no issues are detected.
• C. Utilize zero-copy cloning to create a staging environment of your application, deploy the new version to the clone, and then swap the clone with the production environment for near-instant deployment.
• D. Use the ALTER APPLICATION' command with the 'FORCE option to immediately replace the existing application with the new version, ignoring any potential compatibility issues.
• E. Deploy the updated application directly to the production environment during off-peak hours, relying on thorough testing in a pre-production environment to minimize risks.

解説: (ShikenPASS メンバーにのみ表示されます)

You are developing a Snowflake Native Application with a custom setup script. Within the setup script, you need to perform different actions based on whether the application is being installed for the first time in a consumer's account, or if it's an upgrade from a previous version. Which of the following functions or techniques can you reliably use within the setup script to determine if the current execution is a fresh installation vs. an upgrade? (Assume the app version is maintained and accessible as a variable 'current_version' .)

• A. Use the table 'APPLICATION_REGISTRY' function to check for an existing entry for the application in the current account.
• B. Check if the current version is '1.0', assuming the initial version of the application will always be '1.0'.
• C. Use 'SYSTEM$GET_PREVIOUS_VERSION()' function inside the setup script. If it returns NULL, it is the first install.
• D. Query the 'INFORMATION_SCHEMAPPLICATIONS' view to check if the application name already exists in the consumer's account.
• E. Attempt to read a known persistent setting from the application's metadata. If the setting exists, it's an upgrade; otherwise, it's a fresh install. (Requires metadata setup in a previous version)

解説: (ShikenPASS メンバーにのみ表示されます)