McAfee-Secured Website

Exam Bundle

Exam Code: AZ-104

Exam Name Microsoft Azure Administrator

Certification Provider: Microsoft

Corresponding Certification: Microsoft Certified: Azure Administrator Associate

Microsoft AZ-104 Bundle $44.99

100% Microsoft AZ-104 Guaranteed Success With Testking Training Materials

Get Real AZ-104 Exam Questions & Expert Verified Answers as Experienced on the Real Exam!

  • Questions & Answers

    AZ-104 Questions & Answers

    606 Questions & Answers

    The ultimate exam preparation tool, Questions & Answers mirrors the actual AZ-104 exam while allowing you to hone and measure every aspect of your test-taking ability.

  • AZ-104 Video Course

    AZ-104 Video Course

    132 Video Lectures

    AZ-104 Video Course is developed by Microsoft Professionals to help you pass the AZ-104 exam.

    Description

    This course will improve your knowledge and skills required to pass Microsoft Azure Administrator exam.
  • Study Guide

    AZ-104 Study Guide

    458 PDF Pages

    Developed by industry experts, this 458-page guide spells out in painstaking detail all of the information you need to ace AZ-104 exam.

Microsoft Azure Administrator (AZ-104) Practice Tests

Satisfaction Guaranteed

Money Back Guarantee

Testking provides hassle-free money back guarantee with our products. That is because we have 100% trust in the abilities of our professional and experience product team, and our record is a proof of that.

99.6% PASS RATE
Total Cost: $194.97
Bundle Price: $149.98

Purchase Individually

  • Questions & Answers

    Questions & Answers

    606 Questions

    $124.99
  • AZ-104 Video Course

    Video Course

    132 Video Lectures

    $39.99
  • Study Guide

    Study Guide

    458 PDF Pages

    $29.99

Microsoft AZ-104 Practice Test 68

Question 670

A legacy TCP application behind Azure Standard Load Balancer stores short-lived session state on each worker. The operator wants successive connections from the same client IP to favor the same backend rather than using the default five-tuple distribution. Which rule setting should be evaluated?

  1. Configure the load-balancing rule session persistence to a supported client-IP affinity mode and assess failover implications.
  2. Change only backend health probe frequency and the next request from one client IP will stick to its previous backend.
  3. Enable an inbound NAT rule on every frontend port instead of a load-balancing rule.
  4. Increase the public frontend DNS TTL and DNS caching provides per-client affinity among load-balanced backend VMs.

Explanation

Correct Answer

A:

Client-IP session persistence uses a reduced tuple hash to keep flows from a source address on the same backend while it is healthy, although it is not a substitute for shared session state.

Incorrect Answers

B:

Probe cadence evaluates backend availability and does not set the Load Balancer rule session persistence policy.

C:

NAT rules map specific ports to particular endpoints rather than applying source-IP affinity across a healthy load-balanced pool.

D:

All clients can resolve the same frontend address regardless of DNS TTL. Affinity is controlled by load-balancing rule settings.

Question 671

A Standard Load Balancer has two frontend IP addresses with separate TCP 443 rules targeting the same backend VMs and backend port 443. The application must distinguish which frontend destination IP the client addressed. What rule feature and guest configuration should be evaluated?

  1. Enable Floating IP on each relevant load-balancing rule and configure the backend OS to accept the appropriate frontend destination IPs.
  2. Disable the backend health probe and Azure automatically preserves frontend destination addresses for every rule.
  3. Re-add backend NICs to the same pool and duplicate membership will retain the original frontend destination IP in guest packets.
  4. Create a CNAME alias for the second frontend but leave Floating IP disabled on both load-balancing rules.

Explanation

Correct Answer

A:

For rules that reuse the same backend port across frontends, Floating IP preserves the frontend destination IP in packets. The guest must be configured to accept those IP addresses.

Incorrect Answers

B:

Probe configuration controls backend health, not whether destination IP translation preserves the frontend address.

C:

Backend pool membership does not configure Floating IP behavior. Floating IP is a setting on the relevant load-balancing rules.

D:

DNS aliases do not change destination NAT behavior at the backend; Floating IP must be configured for the specified topology.

Question 672

An operations analyst needs counts of failed requests by HTTP status code from a Log Analytics table. Which KQL operation is the central aggregation?

  1. Filter the KQL query to failure events and use project to select status codes, and projection alone produces grouped counts.
  2. Use take 1 after the failure filter and treat that one event as a representative count for every status code.
  3. Use project-away on the status-code column before grouping, and rely on KQL to infer the removed grouping key.
  4. Use summarize count() by the relevant status-code column after appropriate filtering

Explanation

Correct Answer

D:

Summarize groups log records and calculates counts for each status code, providing a failure distribution.

Incorrect Answers

A:

Project selects columns but does not aggregate events by code; summarize count() by the status code generates the requested distribution.

B:

Take limits returned rows; it cannot calculate a complete per-status-code distribution across the selected log set.

C:

Removing the grouping column discards the dimension required by summarize count() by status code.

Question 673

Application Insights records an event where a dynamic Properties field contains JSON such as {"result":{"errorCode":503}}. An analyst needs to filter requests whose nested errorCode is 503. Which KQL approach is most appropriate?

  1. Group by the complete Properties dynamic JSON object without extracting the nested result.errorCode scalar.
  2. Access the nested dynamic property (for example Properties.result.errorCode), cast it to a scalar integer, and filter on 503.
  3. Convert TimeGenerated to integer and compare it to 503, and timestamps are the nested JSON errorCode field.
  4. Use extend errorCode = tostring(Properties) and compare the entire JSON string to the text 503.

Explanation

Correct Answer

B:

Dynamic property access retrieves the nested value, and toint() or a suitable cast supports a numeric comparison against 503.

Incorrect Answers

A:

Grouping full dynamic values can fragment counts and does not isolate the requested error code property.

C:

Timestamps and dynamic JSON error codes are separate values. Extract and cast the relevant nested property.

D:

The serialized whole JSON object is not equal to the isolated number 503; extract and type the specific property.

Question 674

An application buffers events and sends them to Log Analytics ten minutes after they occur. A query sorted by ingestion_time() produces an order that conflicts with the incident timeline. What should the investigator use?

  1. Sort by Azure ResourceId and its lexical ordering reproduces the original order of event occurrence.
  2. Sort by the alert notification email delivery time and treat it as the timestamp the monitored event actually occurred.
  3. Order by ingestion_time() descending; ingestion time always equals event occurrence time for delayed events.
  4. Order by an appropriate source event timestamp such as TimeGenerated, accounting for client clock quality.

Explanation

Correct Answer

D:

TimeGenerated generally represents event occurrence time, so it is more useful than ingestion time for reconstructing delayed event sequences.

Incorrect Answers

A:

Resource identifiers group events by resource but do not encode the event timestamps needed for chronology.

B:

Notification delivery can lag event generation. Use an event-time field such as TimeGenerated, with awareness of source clock accuracy.

C:

Ingestion_time() tracks when records arrived in analytics; buffered events can arrive later and out of occurrence order.

Question 675

New Azure Monitor Agent installations are healthy but custom logs fail to arrive after the network team restricts outbound traffic. Which dependency is relevant?

  1. Configure VM Insights guest performance collection but omit the required Azure Monitor Agent and DCR association on the VM.
  2. Verify the agent's required Azure Monitor ingestion endpoints are reachable under the supported network design
  3. Change the Log Analytics workspace query permissions while leaving the VM network path to the Azure Monitor ingestion endpoint blocked.
  4. Increase VM managed disk IOPS while leaving the AMA endpoint network route and DCR destination unreachable.

Explanation

Correct Answer

B:

A data collection rule does not bypass network restrictions to the service endpoints that receive telemetry.

Incorrect Answers

A:

The guest-data collection path needs the supported monitoring agent and data collection rule association; enabling a chart alone is insufficient.

C:

Query rights control reading stored logs but do not make the Azure Monitor Agent able to send data through a blocked endpoint.

D:

Guest disk performance cannot correct the agent’s missing ingestion endpoint connectivity or an incorrect data collection configuration.

Question 676

An administrator creates an Azure Monitor metric alert split by VM name dimension across 20 instances. The analyst expects one combined alert only. What should be explained?

  1. Split the alert display into resource tags only, and arbitrary Azure resource tags are automatically supported metric dimensions for every signal.
  2. Use a longer metric aggregation period while keeping all dimensions collapsed into a single alert signal.
  3. A metric alert dimension is only a chart label and never changes which individual time series generate alert instances.
  4. Dimension splitting can evaluate and alert on separate time series rather than only one aggregated fleet value

Explanation

Correct Answer

D:

An alert's scope and dimension selections determine how many independent evaluated series can fire.

Incorrect Answers

A:

Azure metric dimension support is signal-specific; tags do not automatically become selectable time-series dimensions.

B:

Time aggregation alone does not separate individual dimension values for per-instance evaluation.

C:

Dimension splitting can evaluate separate series independently; it can create multiple alert evaluations or fired instances.

Question 677

An Azure Monitor alert processing rule was created in Subscription-A and correctly suppresses notifications for matching resources there. Operations attempts to add a virtual machine in Subscription-B to the same rule's resource scope so that its notifications are also suppressed. The portal rejects that additional target scope. What explains the limitation?

  1. The alert processing rule must be renamed to match the Subscription-B action group before it can target a different subscription.
  2. Each alert processing rule can target resources only within the subscription where that processing rule is defined; create a suitably scoped rule in Subscription-B.
  3. An alert processing rule cannot suppress VM alerts, because suppression applies only to Azure Service Health events.
  4. A processing rule can cross subscription boundaries only after changing the VM's alert severity to Sev0, regardless of resource scope.

Explanation

Correct Answer

B:

Microsoft documents alert processing rule scope as resources, resource groups or a subscription only within the SAME subscription where the rule exists. A separate rule in Subscription-B is needed for Subscription-B resources, with appropriate filters and scheduling.

Incorrect Answers

A:

A name match is not a cross-subscription authorization mechanism. Target resource scope constraints apply regardless of action group naming.

C:

Alert processing rules CAN suppress action groups for fired VM alerts. The special caveat is that they do not affect Service Health alerts; that does not prohibit VM suppression.

D:

Severity is an optional filter on already eligible alerts, not a way to extend a rule's resource scope across subscriptions.

Question 678

Operations is replacing a legacy guest log agent with Azure Monitor Agent on Windows VMs. Which TWO actions are needed to collect selected Windows event channels in a Log Analytics workspace? Select TWO.

  1. Install or enable Azure Monitor Agent on the target VMs.
  2. Enable only Azure VM resource-level diagnostic settings and Windows guest event channels will be delivered without an agent and DCR.
  3. Create and associate a data collection rule selecting the supported event channels and workspace destination.
  4. Give the installed agent Storage Blob Data Reader but omit the selected event-channel data sources and workspace destination.

Explanation

Correct Answers

A:

The AMA component is the guest collection agent responsible for sending selected telemetry.

C:

A DCR declares collected event sources, processing and destination; an association applies it to the relevant VMs.

Incorrect Answers

B:

VM platform diagnostics and in-guest Windows Event Logs have different collection paths. AMA and a relevant DCR are required.

D:

Unrelated storage data permission does not configure AMA guest event collection. DCR association and appropriate sources/destination are needed.

Question 679

A critical metric stops being emitted when a service fails, and the existing alert only fires for high metric values. What monitoring blind spot should be considered?

  1. Grant Log Analytics Reader to the whole team but leave the missing-metrics alert condition defined only as CPU above 95%.
  2. Design a missing-data or availability signal rather than assuming a high-value threshold always catches silence
  3. Increase the high-value threshold further when telemetry stops, and an empty time series will exceed the new threshold.
  4. Increase the existing high-CPU metric alert evaluation frequency to every minute while leaving the rule unable to detect missing data points.

Explanation

Correct Answer

B:

A threshold on reported values may not trigger when there are no data points to evaluate.

Incorrect Answers

A:

Workspace access permissions do not turn a high-value metric threshold into a missing-data detector.

C:

No data is not the same as a high numeric reading. Use a suitable missing-data or availability signal.

D:

Evaluating a high-value threshold more frequently cannot infer an outage from a metric series that has stopped reporting. Use an appropriate absence-of-data or independent availability signal.

Question 680

A security team retained older logs in a low-cost long-term state. An analyst runs an ordinary interactive query that returns only recent data. What should the team check?

  1. Review table interactive retention and the appropriate search/job mechanism for older retained data
  2. Treat archived Log Analytics data as permanently destroyed whenever normal interactive query results are empty.
  3. Grant VM Contributor to the analyst and the role changes archived Log Analytics table retention into interactive retention.
  4. Increase the Log Analytics query time span without using a supported search job or retained data access path for archived records.

Explanation

Correct Answer

A:

Log Analytics data may remain retained beyond the period available to standard interactive queries.

Incorrect Answers

B:

Unavailable in interactive search does not prove there are no retained archive records.

C:

VM resource management permission cannot alter retention availability or turn archival records into ordinary interactive log rows.

D:

Older retained data may require a specific search/job workflow; time range alone does not always restore interactive access.