mcAfee Secure Website
23

Microsoft AZ-104 Bundle

Exam Code: AZ-104

Exam Name Microsoft Azure Administrator

Certification Provider: Microsoft

Corresponding Certification: Microsoft Certified: Azure Administrator Associate

AZ-104 Training Materials $44.99

Reliable & Actual Study Materials for AZ-104 Exam Success

The Latest AZ-104 Exam Questions as Experienced in the Actual Test!

  • 24
    Questions & Answers

    AZ-104 Questions & Answers

    606 Questions & Answers

    Includes questions types found on actual exam such as drag and drop, simulation, type in, and fill in the blank.

  • 25
    AZ-104 Video Course

    AZ-104 Training Course

    132 Video Lectures

    Based on Real Life Scenarios which you will encounter in exam and learn by working with real equipment.

  • exam =30
    Study Guide

    AZ-104 Study Guide

    458 PDF Pages

    Study Guide developed by industry experts who have written exams in the past. They are technology-specific IT certification researchers with at least a decade of experience at Fortune 500 companies.

exam =32

Microsoft Azure Administrator (AZ-104) Practice Tests

Guarantee

Satisfaction Guaranteed

Pass4sure has a remarkable Microsoft Candidate Success record. We're confident of our products and provide no hassle product exchange. That's how confident we are!

99.3% Pass Rate
Total Cost: $194.97
Bundle Price: $149.98

Purchase Individually

  • exam =34
    Questions & Answers

    Questions & Answers

    606 Questions

    $124.99
    exam =35
  • exam =37
    AZ-104 Video Course

    Training Course

    132 Video Lectures

    $39.99
  • exam =36
    Study Guide

    Study Guide

    458 PDF Pages

    $29.99

Microsoft AZ-104 Practice Test 78

Question 770

A company wants to send clients in different countries to specified regional public web endpoints, using a geographic traffic-steering policy above the regional Azure Load Balancers. Which Azure service is suitable?

  1. Add a custom Layer 4 Load Balancer rule to inspect the client country and rewrite website DNS records.
  2. Use an Azure Traffic Manager profile with geographic routing and appropriately configured regional endpoints.
  3. Assign a service endpoint to each country to turn a regional balancer into a global geolocation-aware DNS service.
  4. Configure one internal Load Balancer private frontend and make it publicly geolocation-aware.

Explanation

Correct Answer

B:

Azure Traffic Manager supports DNS-based geographic routing methods to choose from configured regional endpoints for users.

Incorrect Answers

A:

Regional Layer 4 Load Balancer rules do not implement DNS-level country steering of separate geographic endpoints.

C:

Virtual network service endpoints provide selected service traffic paths, not global geographic web endpoint steering.

D:

An internal load balancer has a private frontend and does not become a public global DNS traffic router.

Question 771

A Standard Azure Load Balancer has two backend VMs and a correctly configured health probe. VM-A is marked unhealthy. Packet capture shows TCP connections established to VM-A before the probe failure continue to pass traffic, while all new TCP connections go only to healthy VM-B. Which explanation is correct?

  1. The Load Balancer incorrectly ignores its health probe and continues admitting new TCP connections to VM-A.
  2. Standard Load Balancer removes an unhealthy instance from new-flow selection but permits its established TCP flows to continue until they close or time out.
  3. An HTTP health probe automatically terminates every established TCP flow on VM-A the instant that one probe fails.
  4. A new inbound NAT rule must be configured to migrate existing connections from VM-A to VM-B without a TCP reconnect.

Explanation

Correct Answer

B:

Microsoft documents that when a single backend instance fails health probing, new TCP connections are directed to remaining healthy instances while established TCP connections to the unhealthy backend continue. Existing flow state is not automatically torn down by the probe failure.

Incorrect Answers

A:

The observation explicitly says newly established TCP flows go to healthy VM-B. An unhealthy backend continuing to serve its existing connections is documented Standard Load Balancer behavior, not evidence that it receives new flows.

C:

Health probes control eligibility for new connections; they do not inherently abort every established TCP flow at the moment of probe failure on a Standard Load Balancer.

D:

Inbound NAT rules map a frontend port to a chosen instance, not transparently transplant existing TCP connection state between backend VMs. A failed flow normally requires a client-side reconnect.

Question 772

Backend VMs of a Standard public Load Balancer belong to subnet sn-app. An Azure NAT Gateway with one approved static public egress IP is associated with sn-app, while the Load Balancer has an outbound rule configured with a different public frontend IP. No user-defined route forwards default Internet traffic to a network appliance. Which public source-IP behavior is expected for newly initiated Internet connections?

  1. The Load Balancer outbound rule always wins for members of its backend pool, so every new Internet flow must use the load-balancer frontend IP.
  2. The subnet-associated NAT Gateway takes precedence over the Load Balancer outbound rule; new Internet egress uses an approved NAT Gateway public IP, subject to normal routing.
  3. Both egress methods must alternate source IP for each TCP packet within the same connection.
  4. Azure blocks all outbound Internet traffic whenever a NAT Gateway and a Load Balancer outbound rule coexist on one subnet.

Explanation

Correct Answer

B:

Microsoft documents NAT Gateway as higher priority than a Load Balancer outbound rule for subnet egress. With normal Internet routing as described, new connections use the NAT Gateway public egress address; this is distinct from existing established-flow behavior.

Incorrect Answers

A:

Load Balancer outbound rules can translate outbound connections to the frontend public IP, but NAT Gateway has higher priority on the associated subnet. Backend pool membership alone does not override that precedence for newly initiated Internet flows.

C:

A TCP flow must maintain a coherent source address and transport mapping to work. Azure does not alternate between the two resources for every packet inside a connection.

D:

Both configurations can coexist; NAT Gateway takes precedence rather than creating an automatic conflict that disables all egress. Network filtering and routing still need to allow the intended connection.

Question 773

An application uses Azure Standard Load Balancer with Session persistence set to None. The same client starts several new TCP connections to the frontend using a stable source IP but changing source ports, and some sessions reach different backend VMs. The application can tolerate affinity shared by clients behind the same NAT address. Which Load Balancer setting makes the connections from that source IP prefer the same healthy backend?

  1. Keep Session persistence set to None; the default five-tuple hash guarantees all new connections from a client IP reach the same backend despite source-port changes.
  2. Change the HTTP health probe to HTTPS; a health-probe protocol change automatically enables client-IP affinity.
  3. Enable per-user HTTP cookie stickiness directly on the Layer-4 Azure Load Balancer.
  4. Set the load-balancing rule’s Session persistence to Client IP, using its source-IP/destination-IP two-tuple distribution mode.

Explanation

Correct Answer

D:

Microsoft documents the Client IP two-tuple mode as a source-IP affinity option. It tends to direct successive connections from the same source IP to the same backend, with the trade-off that multiple users sharing a NAT address can concentrate on one backend.

Incorrect Answers

A:

The default None mode hashes the five-tuple including source port. A new connection from the same client with a different source port can be assigned a different healthy backend.

B:

Probe protocol determines backend health eligibility. It does not change the flow distribution hash or provide session persistence for clients.

C:

Azure Load Balancer is a Layer-4 service and does not implement application cookie affinity. Cookies require an appropriate Layer-7 component instead of this transport-level setting.

Question 774

A Standard Load Balancer backend reports healthy to its TCP probe, but one web app accepts the connection then returns a denial from its guest firewall on TCP 8443. The frontend and backend rule are correct. Which layer should be investigated?

  1. The VM guest firewall, application listener binding and effective source/destination NSG rules for the actual client traffic port.
  2. Delete the healthy probe, which guarantees that every client TCP request succeeds.
  3. Increase Storage Blob public access to authorize web connections to the VM.
  4. Change a public DNS TXT record rather than inspect TCP connectivity to the backend.

Explanation

Correct Answer

A:

A successful probe does not establish that client flows to the application listener are authorized; check guest and network filtering at the actual service port.

Incorrect Answers

B:

Removing the probe does not grant the guest firewall’s denied TCP traffic and can prevent reliable backend health selection.

C:

Blob anonymous authorization is separate from VM listener/firewall admission.

D:

TXT ownership records do not determine whether a guest firewall permits the Load Balancer client traffic path.

Question 775

An operator needs the Azure platform-reported Percentage CPU for one VM over the last six hours, displayed at five-minute granularity for capacity planning. No guest process counters or logs are requested. Where should the operator query it?

  1. Inspect the application's relevant response-time metric rather than infer from CPU alone.
  2. Query the VM's platform CPU metric in Azure Monitor Metrics with suitable aggregation.
  3. Configure eligible guest/VM Insights collection rather than assume host metrics include every guest counter.
  4. Account for ingestion delay and verify resource metric availability before claiming data loss.

Explanation

Correct Answer

B:

Azure Monitor Metrics includes the platform Percentage CPU time series for Azure VMs without requiring guest agent setup. Select the VM resource, the Percentage CPU metric, suitable aggregation and a five-minute time grain over the required six hours to obtain the requested series.

Incorrect Answers

A:

An application response-time metric measures user-perceived latency, which may be affected by CPU but is not the VM platform Percentage CPU series. It cannot replace the specifically requested host-reported CPU utilization trend.

C:

VM Insights and guest performance data collection are useful for in-guest memory/process counters but add unnecessary setup for an available VM platform CPU metric. The task does not ask for guest operating-system internals.

D:

Considering ingestion latency helps explain temporarily missing recent samples but does not choose or retrieve the Percentage CPU measurement. The requested data is available directly through the VM Metrics view.

Question 776

A customer-facing API has normal host CPU utilization but its p95 response latency increased sharply after a deployment. Application Insights request telemetry is available. Which signal directly verifies the customer symptom?

  1. The VM platform Percentage CPU metric alone, because normal CPU excludes request queuing.
  2. Azure Monitor alert evaluation frequency, which measures end-user request latency directly.
  3. The application request-duration or dependency-duration percentile and correlated failed requests in Application Insights.
  4. The count of Azure resource groups in the subscription, which tracks response latency.

Explanation

Correct Answer

C:

Request duration percentiles and associated dependency/failure telemetry directly quantify user-experienced latency even when host CPU stays normal.

Incorrect Answers

A:

Platform CPU alone does not measure end-user request duration; a bottleneck may be in an upstream dependency or application path.

B:

Alert evaluation frequency specifies how often a rule runs, not the duration of individual incoming API requests.

D:

Subscription resource-group counts are governance inventory and carry no response-time evidence for the deployed API.

Question 777

An Azure Storage account's transaction failures increased immediately after a firewall/network rules change. Before changing the rules again, an administrator must establish whether the new failures are authorization-related rather than server busy/timeouts, using platform telemetry for the same period. What should the administrator examine first?

  1. Inspect alert time grain, window size and evaluation frequency.
  2. Use suitable numeric aggregation (Average/Maximum) for percentage-type utilization.
  3. Inspect the application's relevant response-time metric rather than infer from CPU alone.
  4. Analyze storage transaction/error metrics with aligned time window and dimensions.

Explanation

Correct Answer

D:

The Azure Storage Transactions metric records successful and unsuccessful requests and exposes a ResponseType dimension, including AuthorizationError, AuthenticationError and ServerBusyError. Compare total transactions split by that dimension over aligned intervals before and after the firewall change. This supports a diagnosis; metrics alone do not modify or repair the storage network policy.

Incorrect Answers

A:

Alert evaluation frequency and time grain affect when an existing alert fires and how its signal is averaged. They do not classify storage transaction responses or distinguish AuthorizationError from ServerBusyError after the firewall change. The storage Transactions metric with ResponseType is the relevant signal.

B:

Average/Maximum are useful for percentage-valued utilization signals, not for distinguishing transaction failure categories. The Storage Transactions metric uses request counts and can be filtered or split by ResponseType and API Name to compare the changed period and response classes.

C:

Application response time reports latency from the client or application perspective and does not reveal which Azure Storage response category increased. The question needs evidence of request failure types after the network policy change, so inspect storage transaction metrics and response dimensions before proposing a remediation.

Question 778

A VM CPU metric in Azure Monitor shows brief utilization spikes that disappear when the chart uses a five-minute Average. The same time grain must reveal both the sustained level and brief peaks. Which configuration is appropriate?

  1. Change the metric namespace to Azure Storage and compare its Total transaction count.
  2. Use the VM availability percentage and ignore CPU altogether.
  3. Select Average only and assume its values must equal the highest one-minute CPU utilization.
  4. Display separate Average and Maximum aggregations of the same Percentage CPU metric over the same five-minute granularity.

Explanation

Correct Answer

D:

Separate Average and Maximum series at the same resource/time grain allow a like-for-like view of sustained load versus peaks.

Incorrect Answers

A:

Storage transactions belong to a different resource workload and cannot reveal the VM CPU peak at the requested granularity.

B:

VM availability and CPU are distinct signals; switching to availability loses the intended peak-versus-average analysis.

C:

Averages smooth short spikes, so one Average series cannot substitute for a Maximum series when peak detection matters.

Question 779

An Azure VM shows host Percentage CPU and disk metrics by default, but the operations team needs the guest operating system's available memory counter. No Azure Monitor Agent or guest data collection rule is installed. Which change supplies the missing signal?

  1. Query the VM's platform CPU metric in Azure Monitor Metrics with suitable aggregation.
  2. Configure eligible guest/VM Insights collection rather than assume host metrics include every guest counter.
  3. Split metrics by supported instance dimension or resource granularity where available.
  4. Analyze storage transaction/error metrics with aligned time window and dimensions.

Explanation

Correct Answer

B:

Install/configure Azure Monitor Agent on the VM and associate a data collection rule selecting the supported guest memory performance counter and destination. VM Insights offers guided onboarding and a predefined guest performance dataset. These mechanisms collect operating-system memory signals absent from default host metrics.

Incorrect Answers

A:

The Azure platform Percentage CPU metric measures host-observed compute utilization, not the guest OS available-memory performance counter. Re-querying CPU cannot create the missing guest memory series.

C:

Splitting an existing platform metric by an available dimension produces more series of that metric, but the memory performance counter was never collected. A dimension filter cannot generate absent guest telemetry.

D:

Storage transaction or error metrics report Azure Storage service operations. They neither measure guest VM memory nor configure the VM agent and collection rule required to report available memory.

Question 780

An Azure Monitor metric alert for VM `vm-ops` evaluates every one minute (`Check every = 1 minute`) but uses a 15-minute `Lookback period` and Average aggregation of Percentage CPU. CPU briefly drops from 92% to 5% for one minute. An operator incorrectly expects the alert's next evaluation to use only the 5% CPU reading. Which configuration detail explains why the alert can remain above its threshold?

  1. The one-minute evaluation cadence checks a rolling 15-minute Average; it does not replace the lookback/aggregation window with a one-minute sample.
  2. Raising the alert's severity from Sev 2 to Sev 1 changes its CPU aggregation window to the duration of the latest sample.
  3. Selecting a different action group causes the metric threshold calculation to ignore all but the newest CPU value.
  4. Changing the alert target to a subscription Activity Log event automatically converts the VM's existing 15-minute CPU average into a one-minute metric.

Explanation

Correct Answer

A:

Metric alert evaluation frequency controls how often the service evaluates the condition. The lookback period independently defines the preceding interval and Average combines its samples, so a one-minute dip might not move the 15-minute result below threshold.

Incorrect Answers

B:

Severity is notification and prioritization metadata. Altering it changes neither the time window over which CPU values are aggregated nor the numerical average used to determine threshold crossings.

C:

Action groups govern what happens after an alert condition fires, such as emailing an on-call address. They do not change CPU measurement, the lookback interval or the aggregation calculation.

D:

An Activity Log alert evaluates resource management events, not a time series of VM platform CPU metrics. Switching alert type discards the requested CPU condition instead of explaining the existing average.