A server with dual 10 GBASE-T RJ-45 adapters is linked to an aggregation switch through a 70-meter horizontal run that was pulled five years ago. Although the link LEDs are lit, the switch shows the connection has renegotiated to 1 Gbps and is logging a high number of CRC errors. Basic continuity testing of the four pairs passes. Which action should you take first to restore the expected 10 Gbps throughput?
Disable autonegotiation and manually set both ends to 10 Gbps full-duplex
Enable IEEE 802.3x flow control on the switch port and NIC driver
Create an LACP bundle using the two NICs to increase aggregate bandwidth
Replace the existing 70-m run with Cat6A shielded twisted-pair that is certified for 10 GBASE-T
10 GBASE-T can operate the full 100 m channel only when Category 6A (or better) cabling is used. Category 6 and especially Cat5e runs longer than about 55 m often fail certification at the higher 500 MHz signaling frequencies required for 10 Gbps, causing down-negotiation and CRC errors. Re-cabling the link with properly shielded, certified Cat6A twisted-pair addresses the underlying physical-layer limitation. Forcing speed/duplex or enabling flow-control does not improve signal integrity, and link aggregation would still rely on the same sub-spec cable.
Ask Bash
Bash is our AI bot, trained to help you pass your exam. AI Generated Content may display inaccurate information, always double-check anything important.
What is the difference between Cat6 and Cat6A cabling?
Open an interactive chat with Bash
What are CRC errors in networking, and why are they significant?
Open an interactive chat with Bash
What role does autonegotiation play in Ethernet connections?