Physical Layer Troubleshooting

In this article we will discuss Physical Layer Troubleshooting, will make brief discussion on Physical Layer Troubleshooting, In last article we discuss about EtherChannel – Increase Bandwidth.

The physical layer transmits bits from one computer to another and regulates the transmission of a stream of bits over the physical medium. The physical layer is the only layer with physically tangible properties, such as wires, cards, and antennas.

Issues on a network often present as performance problems. Performance problems mean that there is a difference between the expected behavior and the observed behavior, and the system is not functioning as could be reasonably expected. Failures and suboptimal conditions at the physical layer not only inconvenience users but can impact the productivity of the entire company. Networks that experience these kinds of conditions usually shut down. Because the upper layers of the OSI model depend on the physical layer to function, a network administrator must have the ability to effectively isolate and correct problems at this layer.

Common symptoms of network problems at the physical layer include:

  • Performance lower than baseline – The most common reasons for slow or poor performance include overloaded or underpowered servers, unsuitable switch or router configurations, traffic congestion on a low-capacity link, and chronic frame loss.
  • Loss of connectivity – If a cable or device fails; the most obvious symptom is a loss of connectivity between the devices that communicate over that link or with the failed device or interface. This is indicated by a simple ping test. Intermittent loss of connectivity can indicate a loose or oxidized connection.
  • Network bottlenecks or congestion – If a router, interface, or cable fails, routing protocols may redirect traffic to other routes that are not designed to carry the extra capacity. This can result in congestion or bottlenecks in those parts of the network.
  • High CPU utilization rates – High CPU utilization rates are a symptom that a device, such as a router, switch, or server, is operating at or exceeding its design limits. If not addressed quickly, CPU overloading can cause a device to shut down or fail.
  • Console error messages – Error messages reported on the device console could indicate a physical layer problem.

Issues that commonly cause network problems at the physical layer include:

  • Power-related – Power-related issues are the most fundamental reason for network failure. Also, check the operation of the fans, and ensure that the chassis intake and exhaust vents are clear. If other nearby units have also powered down, suspect a power failure at the main power supply.
  • Hardware faults – Faulty network interface cards (NICs) can be the cause of network transmission errors due to late collisions, short frames, and jabber. Jabber is often defined as the condition in which a network device continually transmits random, meaningless data onto the network. Other likely causes of jabber are faulty or corrupt NIC driver files, bad cabling, or grounding problems.
  • Cabling faults – Many problems can be corrected by simply reseating cables that have become partially disconnected. When performing a physical inspection, look for damaged cables, improper cable types, and poorly crimped RJ-45 connectors. Suspect cables should be tested or exchanged with a known functioning cable.
  • Attenuation – Attenuation can be caused if a cable length exceeds the design limit for the media, or when there is a poor connection resulting from a loose cable or dirty or oxidized contacts. If attenuation is severe, the receiving device cannot always successfully distinguish one bit in the data stream from another bit.
  • Noise – Local electromagnetic interference (EMI) is commonly known as noise. Noise can be generated by many sources, such as FM radio stations, police radio, building security, and avionics for automated landing, crosstalk (noise induced by other cables in the same pathway or adjacent cables), nearby electric cables, devices with large electric motors, or anything that includes a transmitter more powerful than a cell phone.
  • Interface configuration errors – Many things can be misconfigured on an interface to cause it to go down, such as incorrect clock rate, incorrect clock source, and interface not being turned on. This causes a loss of connectivity with attached network segments.
  • Exceeding design limits – A component may be operating suboptimally at the physical layer because it is being utilized beyond specifications or configured capacity. When troubleshooting this type of problem, it becomes evident that resources for the device are operating at or near the maximum capacity and there is an increase in the number of interface errors.
  • CPU overload – Symptoms include processes with high CPU utilization percentages, input queue drops, slow performance, SNMP timeouts, no remote access, or services such as DHCP, Telnet, and ping are slow or fail to respond. On a switch the following could occur: spanning tree reconvergence, EtherChannel links bounce, UDLD flapping, IP SLAs failures. For routers, there could be no routing updates, route flapping, or HSRP flapping. One of the causes of CPU overload in a router or switch is high traffic. If one or more interfaces are regularly overloaded with traffic, consider redesigning the traffic flow in the network or upgrading the hardware.

Add a Comment

Your email address will not be published. Required fields are marked *