Performance Problems CNX-001 Practice Questions (Page 7)

Questions 31–35

1. 31expert · hard

   A network administrator is troubleshooting a packet loss issue on a cloud network. The administrator has used `ping` and `traceroute` and has identified that the packet loss is occurring at a specific hop. The hop is a cloud provider's internal router. The administrator wants to confirm whether the packet loss is due to the router dropping packets or due to the router's control plane being busy. Which tool or technique would be most effective to make this determination?

   • AUse a packet capture on the source and destination to compare the packets sent and received.
   • BUse `traceroute` with the `-I` option to send ICMP echo requests instead of UDP packets.
   • CUse `mtr` with the `--report` option to get a continuous, aggregated view of packet loss and latency at each hop.
   • DUse the cloud provider's network monitoring tools to check the router's CPU and memory utilization.
   Reveal Solution Discussion Select an answer first
2. 32application · medium

   An organization runs a stateless REST API on a fleet of identical virtual machines behind a cloud load balancer. The API receives a mix of short, frequent requests and occasional long-running report requests. The operations team notices that the long-running requests are consistently being served by the same backend instance, causing that instance's CPU to spike while others remain idle. The load balancer is currently configured with round-robin. Which load balancing algorithm should be used to improve distribution?

   • AIP hash
   • BLeast connections
   • CSource IP affinity
   • DWeighted round-robin
   Reveal Solution Discussion Select an answer first
4. 33foundation · easy

   Which strategy is MOST effective for mitigating packet loss caused by network congestion?

   • AIncreasing the MTU size on the network interfaces
   • BIncreasing the number of DNS queries sent by the client
   • CImplementing Quality of Service (QoS) policies to prioritize critical traffic
   • DDisabling TCP window scaling on the servers
   Reveal Solution Discussion Select an answer first
5. 34expert · hard

   A company has a web application deployed on a fleet of VMs behind a cloud load balancer. The application is a mix of CPU-intensive and I/O-intensive requests. The operations team has configured the load balancer with a weighted round-robin algorithm, assigning a higher weight to the VMs with more CPU capacity. However, they are still seeing uneven resource utilization. Which additional configuration change would most effectively improve the distribution of traffic?

   • AAdd more VMs to the backend pool.
   • BConfigure the load balancer to use least connections instead of weighted round-robin.
   • CIncrease the health check interval to reduce the load on the backends.
   • DEnable session persistence to keep users on the same backend.
   Reveal Solution Discussion Select an answer first
6. 35application · medium

   A company's web application is hosted in a single Azure region. Users in Europe report sluggish performance, while users in the same region as the deployment report no issues. A network engineer runs a traceroute from a European user's workstation and sees the latency increase gradually across several ISP hops before entering the cloud provider's network. The engineer also confirms the application's CPU and database metrics are well within normal ranges. What is the most likely cause of the reported latency?

   • AThe virtual machine's network interface is configured with an incorrect MTU setting, causing fragmentation.
   • BThe application's load balancer is configured with a session persistence policy that is causing uneven distribution.
   • CThe cloud provider's edge router is dropping packets due to a misconfigured access control list.
   • DThe European users are traversing a suboptimal network path with multiple intermediary hops before reaching the cloud provider's edge.
   Reveal Solution Discussion Select an answer first