Shulou(Shulou.com)06/01 Report--

Enterprise local area network due to the complex network environment; the number of terminals, the number of equipment are relatively large; there are often network stutters and other failures. Once the network slows down, in the face of the complex network environment, network management technicians need to quickly and accurately diagnose the problem and solve it. In this article, I will try my best to describe the common causes and diagnostic methods of slow network card changes. Generally speaking, the reasons for network stutter are as follows:

External network reasons: operator line failure, insufficient bandwidth, router software and hardware failure and so on.

Intranet reasons: intranet equipment failure, network settings and other reasons.

Therefore, in the case of network outage and network jam, the first step is to determine whether the problem is in the internal network or the external network, and then make a follow-up diagnosis.

1. First, diagnose through the ipconfig and ping commands.

You should have a Windows computer on hand. Under the dos window, enter ipconfig to view the IP address of the machine "manually set" or "automatically obtained". As shown in the figure:

There are several possibilities:

Both the IP address and the gateway address are correct. (there is no problem, you can proceed to the next step)

The IP address is 0.0.0.0, or 169.254.255.x. (no IP was obtained, indicating that there is a problem with DHCP)

Incorrect IP address, for example: the LAN segment is 192.168.1.x, but the obtained IP address is 192.168.0.x. (indicates that there are privately connected DHCP services, resulting in DHCP conflicts)

When the network is cut off, the first thing is to judge whether there is a problem in the intranet. Generally speaking, it can be judged by the ping intranet gateway. As shown in the figure:

The packet loss rate of the private network should be 0, the wired ping value is generally within 1ms, and the wireless ping value is between 1ms-20ms. If there is packet loss or the ping value is high, then the problem lies in the intranet. Need to troubleshoot internal network cables, loops, switches and other equipment problems. If the wired ping value is normal and the wireless ping value is high (above 50ms), there is something wrong with the wireless device (insufficient signal strength, line interference, insufficient AP performance, etc.).

After the private network problem is eliminated, ping the public network IP address. If it is not available, the problem lies in the router gateway. As shown in the figure below, 114.114.114.114 is a commonly used DNS server. Because 114on the public network, the ping value is higher than that of the private network, generally between 1ms-50ms.

The ping can be connected and the ping value is normal without packet loss, which means that the external network is normal, otherwise the external network is abnormal and it is necessary to troubleshoot the router and wan port line problems.

If the external network can be connected to ping, but still can not be connected to the network, that is generally the problem of dns. It can be judged by the ping domain name.

There will also be viruses and IP address conflicts in the intranet. If you find that the Ip address is incorrect and the private network ping is unstable, you need to consider private network viruses, network loops and other reasons.

two。 Diagnosis of external network problems

If the internal network is normal, then the problem is mostly on the external network line (optical cat hardware) or router. A safer way to judge is to connect the computer directly to the optical cat for online testing. there are the following possibilities:

There is still a problem with the direct access cat network, so it can be confirmed that it is the line. You can report directly to the operator for repair.

The speed of the direct light cat network is normal, and the network slows down after passing through the router, so the problem lies with the router. Consider the aging of router hardware, insufficient performance, poor network cable contact and so on.

3. Summary of some other network problems

There are also some network problems that occur irregularly, and there are generally the following possibilities:

There is a bottleneck in the performance of the router at the peak of the network. It is recommended to check the performance indicators of the router at the peak of the network.

There are occasional broadcast storms. You need to grab the package or use WFilter to determine whether there is a network storm.

The upload bandwidth is insufficient at the peak of the network. For the broadband with unequal uplink and uplink, insufficient uplink bandwidth is also a common cause of failure. You can tell by looking at the total upload bandwidth.

Network management technicians had better prepare some commonly used fault diagnosis software, such as wireshark packet grabbing tools and so on. Or the network health detection plug-in of WFilter (or WSG gateway) can automatically detect ping, arp, tracert and so on, so that most of the network problems can be detected with one click.

You can also view real-time bandwidth usage and so on.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.