Troubleshoot Direct Connect High Latency — mtr and CloudWatch
A network engineer is troubleshooting high latency on a Direct Connect connection. Which TWO actions should the engineer take to diagnose the issue?
Quick Answer
The answer is to review CloudWatch metrics for the Direct Connect virtual interface and use mtr to trace the path. CloudWatch provides latency and packet loss metrics specific to the virtual interface, offering a direct view of performance degradation, while mtr combines traceroute and ping to identify problematic hops along the path, which is critical when troubleshooting Direct Connect high latency. On the AWS Certified Advanced Networking Specialty ANS-C01 exam, this tests your understanding that standard traceroute often fails over Direct Connect due to ICMP filtering by on-premises routers, making mtr a more reliable tool, and that BGP session status or VPC Flow Logs do not measure latency. A common trap is assuming traceroute works universally, but mtr’s ability to use TCP or UDP probes bypasses ICMP blocks. Memory tip: “CloudWatch for the what, mtr for the where”—metrics show the symptom, mtr shows the location.
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
✓
Run a continuous 'mtr' from on-premises to an AWS resource
Running a continuous 'mtr' from on-premises to an AWS resource (C) can identify latency and packet loss along each hop, including the Direct Connect connection. Reviewing CloudWatch metrics for the Direct Connect virtual interface (D) provides latency, packet loss, and throughput statistics. Option A is incorrect because BGP session status only indicates whether the routing session is established, not the latency. Option B is incorrect because VPC Flow Logs show traffic metadata, not latency. Option E is incorrect because traceroute may be blocked by ICMP filtering and is less reliable than 'mtr' for measuring latency over Direct Connect.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Check the BGP session status
Why it's wrong here
BGP status indicates connectivity, not latency.
- ✗
Enable VPC Flow Logs on the VPC
Why it's wrong here
Flow logs show traffic but not latency.
- ✓
Run a continuous 'mtr' from on-premises to an AWS resource
Why this is correct
Identifies latency at each hop.
- ✓
Review CloudWatch metrics for the Direct Connect virtual interface
Why this is correct
Provides latency and packet loss metrics.
- ✗
Run a traceroute from on-premises to an AWS IP address
Why it's wrong here
May be blocked by network devices.
Visual reference
Go deeper
Related to this question
About these practice questions
This ANS-C01 question is part of Courseiva's 1,621-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →
Same concept, more angles
2 more ways this is tested on ANS-C01
These questions test the same concept from different angles. Work through them to make sure you can recognise it however the exam phrases it.
Variation 1. A network engineer is troubleshooting high latency on an AWS Direct Connect connection. The engineer checks the CloudWatch metrics for the virtual interface and sees that 'ConnectionBpsEgress' and 'ConnectionBpsIngress' are both below 50% of the provisioned bandwidth. Which additional metric should be examined to identify potential bufferbloat?
easy- A.CRAM (Challenge-Response Authentication Mechanism) handshake failures
- B.Latency metrics from Amazon CloudWatch
- ✓ C.PauseFrames (802.3x flow control frames)
- D.CRC errors on the physical connection
Why C: (PauseFrames) is correct because PauseFrames indicate that the link is being flow-controlled due to congestion, which can cause bufferbloat. High pause frame counts suggest the local device is forcing the remote device to pause transmission, leading to increased latency. Option A is incorrect because CRAM is a security protocol, not a network metric. Option B is incorrect because detailed latency metrics are not available for Direct Connect virtual interfaces. Option D is incorrect because CRC errors indicate physical layer issues, not bufferbloat.
Variation 2. A network engineer is troubleshooting high latency on a Direct Connect connection. The engineer wants to use monitoring tools to identify the source of the latency. Which two AWS services can provide metrics and logs to help diagnose the issue? (Choose TWO.)
medium- A.AWS CloudTrail
- B.AWS Config
- ✓ C.Amazon CloudWatch
- D.AWS Trusted Advisor
- ✓ E.VPC Flow Logs
Why C: The correct answers are C (Amazon CloudWatch) and E (VPC Flow Logs). CloudWatch provides metrics for Direct Connect, including connection state, BGP status, packet loss, and latency. VPC Flow Logs capture IP traffic information that can be analyzed to identify latency patterns. Option A (AWS CloudTrail) is incorrect because it logs API calls, not network metrics. Option B (AWS Config) is incorrect because it tracks configuration changes. Option D (AWS Trusted Advisor) is incorrect because it provides recommendations, not real-time metrics or logs.
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This ANS-C01 practice question is part of Courseiva's free Amazon Web Services 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 ANS-C01 exam.