You save $69.98
SY0-701 Premium Bundle
- Premium File 1044 Questions & Answers
- Last Update: Nov 18, 2026
- Training Course 167 Lectures
- Study Guide 1003 Pages
You save $69.98
A municipal water facility calls its process network 'air-gapped' because no production Ethernet link crosses to the corporate LAN. Operators nevertheless import spreadsheet reports from office USB drives weekly. What limitation must be recognized?
A drive can carry files, code or misconfiguration into an otherwise disconnected network. Controlled media, dedicated transfer hosts and logged authorization reduce that exposure. The utility should test that the process network remains isolated while allowing only validated business transfers. The transfer process should include approved content types, recorded custodians and malware inspection before any report reaches the process environment.
Physical network separation stops certain packet paths but does not prevent people moving files or devices between networks. Malicious media can introduce code or sensitive information across the boundary. Air-gap assurance must account for human and removable-media workflows.
An office scan can reduce some known malware risk, but transferring removable media across the boundary still creates a bridge. Macro, parser and device-level threats may survive ordinary scanning, and processes on either side could handle the files unsafely. A formal controlled-transfer mechanism must include the process-side threat and provenance checks.
A scan at a transfer station can reduce risk but does not establish that imported files are inert, especially if the receiving software executes macros or contains exploitable parsers. Physical disconnection from the corporate LAN is incomplete protection when staff repeatedly move data on removable media. The facility needs controlled media provenance, content validation and safe import procedures instead of assuming a scanner alone closes the gap.
An engineering campus segments a sensitive test lab using SDN overlays. A design memo claims that an SDN deny policy is literally equivalent to disconnecting every physical interface. Which response is correct?
An SDN policy can be useful for segmentation and microsegmentation, but it remains a logical control dependent on software and connectivity. A physically disconnected interface provides a different failure model. The lab should align the required assurance level with actual isolation evidence and management-plane risk. An SDN control-plane failure or unauthorized policy change could alter connectivity without a technician moving a cable. The design should define which management principals may change the overlay and how forwarding rules are validated against the intended isolation level.
A controller policy is a logical traffic restriction subject to configuration, forwarding bugs and controller compromise. It does not remove the physical connectivity that makes an air gap distinct. The campus may adopt SDN microsegmentation, but its assurance documents must not describe a connected network as physically disconnected.
Controller redundancy improves availability of management decisions but does not prove hardware separation or protect against every misapplied forwarding rule. A compromised switch or inconsistent overlay can still carry traffic despite healthy controllers. Physical air-gap claims need verification of actual disconnected interfaces and pathways.
Central policy can reduce lateral movement, but an allowed host, overly broad service rule, or compromised forwarding device can still become a crossing path. The lab needs tests of effective flows and administrative boundaries. Logical segmentation does not promise the absolute disconnection asserted in the design memo.
A clinic needs its patient portal available during failure of one server but can afford only two compute nodes. Which arrangement most directly avoids one-node hardware failure causing a complete outage?
Two appropriately configured nodes can preserve availability when one stops, subject to shared-storage and network dependencies. Failover testing should confirm the clinical service recovers within its target, not just that a standby machine exists. Separate power paths reduce common-mode risks when feasible under budget. The reviewer should check for common power, storage and load-balancer failures that could still make both nominally redundant nodes unavailable.
Configuration backups help reconstruct a failed system, but they do not supply a running compute node or meet a short portal-availability objective by themselves. A restore may require an engineer, replacement capacity and testing, making this a recovery plan rather than node-failure tolerance.
Two machines that share one unprotected power or network dependency may fail together. Without failover, a spare may not take over automatically. Redundancy must be engineered and tested rather than inferred from the number of physical boxes.
Client retries can help with intermittent requests when a service remains available, but they cannot make a destroyed or offline server process requests. Without a second functioning instance and a path for traffic to reach it, a permanent node failure remains a complete outage.
An analytics service handles 2,000 requests per minute normally and 20,000 during financial reporting week. Its storage and database tiers cannot scale as quickly as its web servers. Which evaluation best tests scalability?
Scaling one layer cannot compensate for a saturated database or shared queue. End-to-end tests reveal capacity and failure behavior at realistic demand. The finance team should measure latency, errors and security-control effectiveness as load increases, then address the actual limiting tier. A capacity test that monitors only the web tier may miss transaction delays caused by locked database connections or exhausted worker pools.
A static page can be served quickly while data queries and authentication are overloaded. Testing only an easy endpoint would hide the bottleneck behind actual business reports. The performance model must include representative transactions.
More user interface elements do not create compute, database or network capacity. The system could become slower as additional dashboards request the same data. Architecture should be assessed with measured throughput and resource metrics.
Skipping authentication would create a security hole without solving underlying database scaling limits. Demand planning must preserve access control and resource protection. Cutting critical security checks is not an acceptable substitute for capacity engineering.
A small business can keep its customer site online during a minor server restart but cannot recover for two days after an entire location is lost. Which distinction should the resilience assessment emphasize?
Availability measures whether service is accessible, while resilience encompasses absorbing disruptions and recovering after failures. The two-day regional outage reveals a continuity gap despite good everyday uptime. The business should define RTO/RPO and test an alternate location or other realistic recovery path. An agreed recovery target is meaningful only if an exercise demonstrates that staff, data and infrastructure can restore the service after a site-wide loss.
Historical uptime can describe how often the website was reachable, but it does not establish whether the system can recover within a defined time after losing a whole facility. A company can report excellent routine uptime yet remain down for days after regional failure. Resilience assessment needs tested major-incident recovery behavior and objectives.
Local redundancy may protect against one server failing, but it shares power, location and networking risks with the rest of the facility. The scenario specifically describes a two-day outage when the location is lost. The assessment must distinguish routine component availability from recovery of entire-site dependencies.
Automated health checks and restarts help with transient application failures, but cannot bring back a destroyed building, power supply or local storage. The two-day site-loss recovery gap remains. A meaningful resilience plan includes restoration capability, alternate resources and tested recovery objectives rather than only faster in-place restarting.
A company places traffic-inspection software on battery-powered edge gateways. Under heavy load, packet loss rises and available power drops below the safe reserve for sensor operation. What placement decision is most appropriate?
Shifting deep inspection to an upstream gateway or aggregation service can preserve visibility without exhausting low-power devices. Lightweight local enforcement can still restrict unauthorized connections. The deployment should be tested for detection gaps, network latency and sensor uptime under worst-case load. An upstream inspector should receive the necessary packet metadata without requiring the battery-powered device to duplicate resource-intensive analysis on every request.
More processing on an already saturated edge device will worsen packet loss and may harm the sensors it supports. Security controls need to operate within the platform's real resource envelope. A measured architecture change is safer than escalating the overload.
More power may reduce one resource constraint, but the stated problem includes packet loss and a time-sensitive sensing workload under inspection load. Without performance testing or placement redesign, simply increasing battery size could leave the gateway unable to meet its core availability requirements.
Removing authentication would create a large access-control weakness and does not directly solve inspection placement. The company can retain low-overhead identity and packet-filter controls while relocating expensive analytics. Broad security removal should not be the first response to capacity pressure.
A retailer has a container-based order service with an unpatched dependency whose vendor has published a fix. Operations argues the architecture is safe because the service is replicated across three nodes. Which decision is most defensible?
A vendor fix offers a direct way to correct the vulnerable dependency, subject to testing and controlled rollout. Container replication may reduce downtime during deployment but leaves the defect present until images are updated. Risk acceptance should consider exposure and patch evidence, not replication as a substitute.
Replication normally improves service availability; it does not remove the same vulnerable dependency from each image. An attacker may need to compromise only one reachable instance to gain a foothold or data access. The retailer should test and deploy the available vendor fix while coordinating safe rollout across nodes.
A WAF can mitigate selected web exploits but does not cover every vulnerable library behavior or internal request path. An immutable image simply preserves whatever code was built into it until replacement. The published supported fix and testing remain necessary, with compensating controls for the transition window.
A valid vendor fix reduces security exposure but may introduce compatibility or startup issues in a specific deployment. Taking down all replicas together removes the redundancy that protects availability. A staged rollout with acceptance checks and rollback delivers the patch without turning a manageable vulnerability into an avoidable outage.
A company hosts a public API and an internal billing database in one cloud virtual network. Firewall controls are applied only within each subnet, leaving the connection between them broadly permitted. Where should the key control sit?
Boundary enforcement restricts requests according to source identity, destination and permitted service behavior rather than merely subnet membership. The company should measure actual API-to-database dependencies, then allow only necessary communications. This limits lateral movement if a public-facing component is compromised. A test from the compromised public API subnet should confirm that arbitrary queries to the billing database are rejected while legitimate application transactions still succeed.
A WAF can filter selected hostile HTTP requests at the public interface and is useful defense in depth, but it does not enforce the trust boundary between a compromised API workload and billing storage. Arbitrary lateral database connections can still traverse the broadly allowed internal route. The company needs explicit destination, identity and port restrictions at the service-to-data boundary.
Encryption at rest can protect stored media under appropriate key custody, but it does not prevent an API process from making live connections to the billing service. A compromised workload may still obtain decrypted records via allowed application paths. The issue is excessive cross-zone reachability, so network and application authorization must limit specific API-to-billing flows.
Passive detection can show suspicious lateral traffic and preserve investigation evidence, but it does not prohibit the traffic from reaching sensitive billing systems. The stated problem is a permissive connection between security zones. An enforceable allow-list at the crossing, supported by detection, addresses the direct access risk rather than relying on alerts after exposure.
A warehouse runs a visitor-facing delivery application and a database holding employee identifiers. Developers want to place both in the same flat subnet for easier troubleshooting. Which architecture better limits risk?
Tiered security zones separate the external attack surface from the sensitive data store. A defined application path can preserve delivery workflows while limiting direct database reachability. The warehouse should test unauthorized connections from the public zone and ensure the app's identity has least-privilege query access. Separate routing zones reduce direct reachability, but the authenticated application still needs minimal database permissions and prepared access paths. Verify both internet-to-database denial and allowed delivery queries, because a firewall diagram alone does not prove enforcement.
Publishing the database adds an unnecessary entry path and broadens the exposed surface. Application authentication does not automatically protect a separate database listener. The database should not be independently reachable from external clients without a defined need.
A login page on the delivery application does not by itself protect a database service from direct network reachability or compromise of a public frontend. The employee records need an independently enforced trust boundary and narrowly authorized service-to-database traffic. Publishing the backend through the same ingress path would increase access surface rather than isolate sensitive data from visitors.
Logs help demonstrate and investigate which application flows crossed the boundary. Sensitive content can be minimized in monitoring, but disabling it removes important security evidence. The proposed architecture should include appropriate telemetry as well as enforcement.
A factory designs two gateway modes: 'fail open' allows traffic if policy enforcement fails; 'fail closed' blocks traffic until the gateway recovers. For a non-safety-related confidential research interface, which behavior better preserves access confidentiality during enforcement failure?
Fail closed denies access when the enforcement decision cannot be trusted, helping preserve confidentiality. An approved break-glass route can be designed for genuine emergencies without making all normal traffic unrestricted. The factory should test recovery time and ensure fail-closed behavior does not unintentionally disable safety systems. The approved emergency exception must be narrow enough that a malfunctioning enforcement service cannot turn normal research access into unrestricted connectivity.
Fail open can preserve connectivity but may grant access without normal policy checks, conflicting with the stated confidentiality priority. It may be appropriate for other safety or availability constraints, but not as an automatic choice for sensitive research access. The decision must reflect the protected asset and failure impact.
A previously authenticated client may have a stolen session, changed risk or authorization limited to one resource. Continuing unrestricted access during policy failure turns temporary availability into implicit trust and can expose confidential research interfaces. A controlled break-glass process should instead explicitly document narrow access and evidence requirements while the ordinary gateway fails closed.
An indefinite cached allow can outlive token revocation, role changes or a newly detected compromised endpoint. The interface is confidential and not safety-critical, so access should be blocked when the protective decision cannot be enforced. Emergency access, if needed, should be separately scoped, authenticated and logged instead of silently extending prior grants.
A municipal emergency service requires response times under ten milliseconds. Security architects propose a new inline prevention appliance that adds an untested fifteen milliseconds during peak loads. What should be decided before deployment?
A representative latency test can reveal whether inspection can be optimized, moved out of line, or replaced with complementary controls. The municipality should maintain its ten-millisecond service objective while assessing actual threat coverage. The final design must be verified at peak traffic, not just during idle benchmarks.
A prevention demonstration establishes that selected attacks can be blocked, not that critical emergency transactions meet the documented ten-millisecond response requirement. The device already showed an untested fifteen-millisecond peak-load delay, which is a serious deployment risk. The team must first test realistic traffic or change placement and design, with documented application-owner acceptance of performance.
A security engineer cannot safely redefine a clinical or emergency-service latency objective merely to accommodate one product. The ten-millisecond constraint comes from the application’s operational requirements and needs accountable owner review. A better decision is testing alternatives such as out-of-path detection, faster inspection, or other placement that provides protection without violating the known response-time limit.
Bypassing inspection might reduce delay but also removes the proposed prevention exactly when emergency systems are busy. The decision must be supported by measured latency budgets, failure modes and alternate controls, not an untested always-bypass exception. A security design should meet the response-time target while retaining necessary protection and documenting any approved risk trade-off.
Please check your mailbox for a message from support@examlabs.com and follow the directions.