Question

You manage an Elastic Job agent that runs a nightly index rebuild against a target group containing 40 Azure SQL databases spread across three logical servers. Each database has a different name, and you need the same T-SQL job step to execute successfully against every database without hardcoding names. The job step currently fails on some databases because they use different schema names for the target table. You want the job step to reliably enumerate and act on each database's local objects at runtime. Which characteristic of Elastic Jobs must you rely on to make the single job step work across all targets?

Accepted Answer

Elastic Jobs automatically connect to and execute the job step's T-SQL independently in the context of each target database, so the script should use dynamic SQL or system catalog views to resolve local objects at runtime.

Answer

Elastic Jobs execute the T-SQL once on the job database and replicate the resulting rows to each target, so you must design the script to write to a shared output table.

Answer

Elastic Jobs require a separate job step per target database, so you must generate 40 job steps that each specify the database name and schema explicitly.

Answer

Elastic Jobs run the T-SQL in the master database of each logical server and use three-part names to reach each target, so the script must fully qualify every object with the database name.