Download Latest Development-Lifecycle-and-Deployment-Architect Questions

 Extracting each permission set, committing and merging to source control, and running through CI checks is the best strategy to bring the issue of failing SOQL statements to the developer’s attention earlier. This way, the developers can review the changes made by the administrators, and ensure that they do not break any existing functionality or security rules. Asking administrators to only make changes to profiles instead would not solve the problem, as profiles can also affect SOQL queries. Creating a sandbox refresh strategy to ensure each sandbox is refreshed every day would not prevent the issue from happening, as the changes made by the administrators would still be propagated to the sandboxes. Advising developers to switch to SOSL queries that are more robust instead would not address the root cause of the issue, which is the lack of visibility and control over the permission sets.

The Tooling API is the best choice for the team that wants to use SOQL queries to retrieve and make changes to smaller pieces of Salesforce metadata through their Java application. The Tooling API provides access to metadata objects such as Apex classes, triggers, workflows, validation rules, custom objects, and fields. It also supports CRUD operations, as well as executing anonymous Apex code. The other APIs are not suitable for this use case, as they either deal with larger metadata components (Metadata API), user interface components (User Interface API), or data records (Any Salesforce API).

 Executing all test classes in sandboxes can help identify and fix any errors or failures before deploying to production. Executing selected test classes in production can help reduce the deployment time and avoid running unnecessary tests.

 The two deployment tools that will support the request are the ANT Migration Tool and VS Code with Salesforce Extension. These tools allow the developers to deploy metadata components from one Salesforce org to another, and also to maintain and modify the code locally in their own machines. Change Sets and Developer Console are not suitable for this scenario, because they do not support deploying to multiple orgs or working offline.