mediumMultiple Select
PDE Practice Question: A Dataflow streaming job is processing data from…
A Dataflow streaming job is processing data from Pub/Sub and writing to BigQuery. The job is stuck with the message 'No progress has been made' for several minutes. Which TWO actions should the team take to troubleshoot and resolve the issue? (Choose TWO.)
⚠ Common exam trap
Google Cloud often tests the misconception that increasing resources (like disk size) or restarting the pipeline is the default fix, when in reality the first step is always to inspect logs to understand the failure mode.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Examine the worker logs in Cloud Logging for any error messages or exceptions.
Option C is correct because when a Dataflow streaming job reports 'No progress has been made,' the first troubleshooting step is to inspect the worker logs in Cloud Logging, which surface exceptions, stuck DoFns, BigQuery write errors, or Pub/Sub backlog issues that explain the stall. Option E is correct because enabling Dataflow Streaming Engine moves pipeline state (such as timers and stateful processing) off the worker VMs and into the Dataflow backend, reducing worker CPU/memory pressure and often resolving stalls caused by state-heavy or resource-constrained workers. Option A is incorrect because updateCompatibility is not a valid Dataflow pipeline option for resolving a stuck job. Option B is incorrect because increasing persistent disk size addresses disk I/O capacity, not the typical causes of a streaming job making no progress. Option D is incorrect because force-stopping and redeploying with --update does not diagnose the root cause and can lose in-flight state without first examining logs.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Set the updateCompatibility flag to true and restart the pipeline.
Why it's wrong here
updateCompatibility controls pipeline update behaviour during job replacement, not the watermark stall causing 'No progress'. It is tempting because pipeline updates can appear stuck, and the flag is genuinely needed when upgrading a running job whose transforms changed incompatibly, but it cannot unblock a streaming source or sink.
- ✗
Increase the persistent disk size for all workers to reduce I/O contention.
Why it's wrong here
Persistent disk size affects shuffle and temporary storage capacity, not the watermark advancement that 'No progress' reports. Enlarging disks is genuinely useful when workers hit disk exhaustion or shuffle spill errors, but the stalled watermark here stems from the source backlog, sink write failures, or windowing, none of which disk capacity addresses.
- ✓
Examine the worker logs in Cloud Logging for any error messages or exceptions.
Why this is correct
Worker logs in Cloud Logging capture exceptions, permission errors, and BigQuery write failures that stall a streaming pipeline, directly explaining the 'No progress' state. Examining them satisfies the troubleshooting requirement by surfacing the underlying error before any configuration change.
- ✗
Force stop the pipeline and update it with a new version using the --update flag.
Why it's wrong here
Force-stopping discards the running pipeline's state and cannot be combined with --update, which requires a live job to replace; restarting also loses the stuck watermark evidence needed for diagnosis. Draining and updating suits planned code deployments, not troubleshooting a stalled streaming job.
- ✓
Enable Dataflow Streaming Engine to move state to the backend and reduce worker load.
Why this is correct
Streaming Engine moves pipeline state and shuffle to the Dataflow backend service, so workers no longer hold state in memory or on persistent disk. This reduces worker load and memory pressure, which is a common cause of stalled watermark progress and the 'No progress has been made' message.
Visual reference
Go deeper
Related to this question
About these practice questions
One of 747 original PDE practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This PDE practice question is part of Courseiva's free Google Cloud certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the PDE exam.