Data-Engineer-Associate - Working with Lakeflow Jobs (16% of the exam) - Section 4.3

Implement job schedules using Lakeflow Jobs with an understanding of trigger types (scheduled, file arrival and table update).

Set up scheduled (cron) triggers, file arrival triggers on a storage location, and table update triggers, and state what starts a run under each. Recognise concurrency settings and what happens when a run is still in progress at the next trigger.

scheduled triggerfile arrival triggertable update triggercron schedulemaximum concurrent runs

Practice question for this objective

Free sampleWorking with Lakeflow Jobsmedium

A partner writes a handful of CSV files into a cloud storage location at irregular times between 06:00 and 22:00, and on some days writes nothing at all. A team currently runs a Lakeflow Job on a cron schedule every fifteen minutes, and most runs find no new data to process. The team wants a run to begin shortly after files land and no run at all on days when nothing arrives. Which statement correctly describes what a file arrival trigger provides here?

AIt subscribes the job to cloud storage notification events, so a run begins within a second of each object being written and the storage location is never polled by the job.
BIt checks the configured external location or volume path on a short interval and begins a run when new files are found there, so stretches of time with no arrivals produce no runs at all. Correct
CIt begins one run of the job for every individual file written to the location, so a drop of twelve files inside one minute produces twelve separate runs of the same job.
DIt replaces the cron schedule with a single daily check that begins one run and processes whatever files have accumulated in the location since the check on the previous day.

A file arrival trigger checks a configured storage location on an interval and starts a run when new files appear, avoiding empty scheduled runs. The file arrival trigger evaluates the configured external location or volume path on a recurring interval and starts a job run when files that were not present at the previous check are found. Because the trigger is driven by the presence of new files rather than by a clock time, intervals with no arrivals start nothing, which removes the empty runs a frequent cron schedule produces.

Why A is wrong: Notification based discovery is how Auto Loader can find files inside a running query, which makes this feel familiar, but the job level file arrival trigger works by checking the configured location on an interval rather than by subscribing the job itself to storage events.

Why B is correct: This is the behaviour of the file arrival trigger: the location is checked on an interval, a run is started once new files are seen, and quiet periods simply produce no runs, which is exactly what the team asked for.

Why C is wrong: A candidate who reads the word arrival as meaning per object can land here, but files seen together are handled by a single triggered run, and per file runs would make bulk drops unworkable.

Why D is wrong: This describes an ordinary daily scheduled trigger wearing a different name, and it would leave the team with the same long wait between a file landing and the run that reads it.

See more Data-Engineer-Associate practice questions, answers explained.

Exam traps in Working with Lakeflow Jobs

Answers that look right on this material and are not. Each one is a distractor from a different question in the Data-Engineer-Associate bank for this domain.

Set the job's maximum concurrent runs to 3, so the overlapping starts are admitted in a controlled number and finish without contending for the same compute.
Why it is wrong: Raising the ceiling does bound the overlap, which sounds like control, but it still permits three runs to append to the same table at once and so leaves the duplication the team is trying to remove.
Pausing the schedule on the job also cancels whichever run is currently executing, so a paused job can never be left with a partially completed transformation behind it.
Why it is wrong: Tempting because pausing feels like a stop button, but pausing only suppresses future triggers. A run already in flight carries on to its own success or failure.
The trigger reads each newly arrived file and appends it to the target Delta table itself, so the job no longer needs a task that opens the landing path and parses the records.
Why it is wrong: Tempting because the trigger is configured with a storage path, but a trigger only starts a job run. Loading the files remains the work of the job's own task, whether that task uses Auto Loader or COPY INTO.

Examworthy is not affiliated with or endorsed by Databricks. Original, blueprint-aligned practice material only.