Router Network Interface resolution Daquan 3

(17) queue messages

For the output and input queues, a number is displayed in the form of m/n, followed by the number of groups lost because the queue is full. Here, the m value is replaced by the number of groups in the queue, while the n value is replaced by the n value to indicate the maximum queue size. By checking the number of lost groups and the relationship between m and n over a period of time, you can determine whether to adjust the queue length of a specific interface to reduce the number of lost groups. However, the media and usage level connected to the interface should also be considered to determine whether debugging of the output queue length is beneficial. Medium with high usage is most likely to cause loss of groups in the queue: When the router transmits data, it will encounter difficulties, resulting in queuing of the output group, which in turn leads to when the output queue is full, in addition, when other groups arrive, the group is lost when it is transmitted to the media through the interface. On the input side, the large ratio of the lost group to m to n indicates that the router is busy with other work and cannot process the incoming group in due time. If the next situation lasts for a long time, it usually means a more powerful router is needed to meet the work needs. In general, this situation can be observed through a large number of lost groups in the Inbound direction of many router interfaces.

In the show interfaces above, the queue information field value shows that no group exists in any queue currently. In addition, although the output queue is full, 63 groups are lost, but no groups are lost due to the input queue. The latter is a common situation because most routers (unless excessively configured) should not have problems in processing incoming data.

(18) 5-minute I/O Rate

The next field shows the average number of digits sent and received through the interface and the average number of groups in the previous 5 minutes. Several factors must be taken into account when interpreting the data displayed in this field. First, you must consider the operation mode of the interface and the network configuration of the interface. For example, if the interface is a LAN interface, it can be run in the chaotic mode, so that every investigation on the Qu LAN can be run in non-chaotic mode, that is, it only reads the broadcast consumer and directly delivers it to the consumer of the interface.

If the port is in the disordered mode, read all groups and provide a method to test the data flowing in the network. If the interface is not in a chaotic state, it only feels like the traffic she sends and receives, which may only occupy a small part of all traffic on the network.

Considering the network configuration, if an interface is connected to a LAN with only one station, such as a WEB server, all traffic will flow through the router interface. This means that you can obtain a relatively accurate method for testing network activity without considering the interface mode.

Another factor to consider is the fact that the 5-minute I/O rate represents the power average of the 5-minute time constant. Therefore, any 5-minute I/O rate is the approximate value of the traffic per second during this period. However, the average value generated by the four 5-minute time span is less than 2% of the instant rate of the unified traffic within 20 minutes.

Because the group length is variable, the bit rate per second is generally more useful than checking the activity on the interface from the transport media perspective. In the preceding example, the input rate of 1/6 BPS is about of the interface running rate. You may wonder why the input rate is about an order of magnitude higher than the output rate of the interface. The answer is interface connection. In this specific vro environment, the Ethernet interface is connected to a 10BASE-TLAN with only one other station (that is, the company's WEB Server. Web page requests flow in the form of a uniform resource locator (URL), and the response to URL requests is a web page; this explains why the input and output traffic levels are not proportional. Now we know the I/O rate for 5 minutes. Next we will introduce the input and output information of specific groups that can be displayed for a specific interface.

(19) group and byte Input

This field indicates the total number of non-error groups received by the vro. Second, it also indicates the total number of bytes of the non-error group received by the router.

If you divide the number of nodes by the number of groups, you can obtain the average length of the Group in bytes. This information can be used to provide a general representation of the flow type on the interface. For example, a relatively short group usually transmits interactive query/response traffic, while a relatively long group usually transfers files including web pages and images contained on most of these pages.

(20) No Buffer

The no-buffer field indicates the number of groups received by the interface that have to be discarded due to the lack of buffer space on the router. Do not mix the buffer space with the interface's internal buffer. When there is a continuous "no buffer" situation, it usually means that the router needs more memory. However, if the no buffers value is set regularly, it may be caused by a broadcast storm on the LAN or a noise attack on the serial port. You can check the next field to determine whether the cause of no buffer value is caused by a broadcast storm.

(21) broadcast received

This field indicates the total number of broadcast or multicast groups received by the interface. Note that many broadcasts are part of the natural communication process. For example, the ARP used to resolve a layer-3 IP address to a layer-3 Mac address depends on the issue of a broadcast, in order to query each station of the LAN of the layer-7 address related to the required layer-2 address, so that the system can correctly form a detection to pass the group. Similarly, in the Novell IPX environment, the server is grouped every 30 s of broadcast service declaration protocol (SAP. These define the services provided by the server.

If you are in a strict IP environment, you are more likely to obtain some broadcasts from ARP requests. If you have a time-based application, you can set a fixed item as the ARP cache for the application that has been running for the time, so that you can solve two problems with one action. This not only prevents the router from performing ARP operations, but also allows the parsing process to occur by checking the memory, which is much faster than waiting for the broadcast response. Because data traffic is interrupted during ARP broadcast, reducing ARP broadcast can improve the interface information transmission function. Because the ARP table is maintained inside the vro.

(22) Runts

Runt is the term of an error scenario. The Group length associated with it is smaller than the minimum length related to a protocol. In an Ethernet environment, the minimum group length is 64 bytes on the adapter and 72 bytes on the LAN. Therefore, if an interface receives an Ethernet group of less than 72 bytes, it will be an error and the group will be discarded. In general, a conflict can cause Runt generation, and a faulty adapter can also cause this situation.

(23) Giants

Another error occurs in Giants. It indicates that the length of the group exceeds the maximum length of the Protocol. In an Ethernet environment, the maximum group length of the adapter is 1518 bytes, and the maximum group length of the adapter flowing in the network is 1526 bytes. Therefore, groups with a length of more than 1526 bytes (including the forward code and the start delimiter field) are considered as Giant. Such groups are also discarded, and the number of Giant groups indicates the number of discarded groups due to this situation. The common cause of the Giant group is a lag conflict or an adapter failure.

(24) Throttles

Although this rarely happens, if the router detects a buffer or processor overload, it will turn off its receiver. This situation is called Throttles, but it is not actually a communication problem. On the contrary, it is a router function problem and requires you to check the status of the system buffer and processor. If the show interfaces command indicates a large number of "unbuffered" and Throttle, it usually indicates that you should consider adding memory to the router. [End]
Article entry: aaadxmmm responsible editor: aaadxmmm