SCO TCP/IP network management-tool Introduction

SCO TCP/IP network management-tool Introduction
After the last six articles, we should introduce the TCP/IP management a lot. Next I will take the end and introduce some TCP/IP tools in UNIX, I hope my article will be helpful to you. I will go back to the SCO User Guide to give some friends who are just getting started a rough idea of SCO OPENSERVER.
Excessive network load will also lead to a decrease in network performance. Remember the following figures: the theoretical maximum transmission rate of DETHERNET is 10 Mbps. The theoretical maximum transmission rate of Fast ETHERNET is 100 Mbps. The theoretical maximum transmission rate of the Token Ring is 4 Mbps. The theoretical maximum transfer rate of Token Ring 2 is 16 Mbps.
There are several tools that can be used to diagnose network performance.
Ping-f: The usage is as follows: 1. It can only be used when the user logs on as root. 2. Transmit a large number of ICMP packets to the target system. 3, used to determine whether the NIC or CPU can be controlled to load. 4. shared with the-c option, you can increase the ICMP packet size to increase the load on the remote host. Generally, the load capacity of a remote host can be measured in the following steps: 1. Run the netstat command and select appropriate parameters for the system to display the running status of a certain performance. 2. record the data you need. 3. Run the ping-f-s packet-size remote-host command. packet-size indicates the size of the ICMP packet sent. Remote-host is the host that receives data packets. 4. Run the netstat command again and use the same ELE. Me parameter as in step 1. 5. Compare the system data in step 2 and step 4.
Netstat-I can display data related to each network interface and help you diagnose network load problems. Run this command. The system displays two columns of data: Collis and Opkts. Divide the data in COLLIS by the data in OPKTS and multiply it by 100. The resulting number is the proportion of the data packet sent by the machine that caused the data error. If the number exceeds 10%, the system load is too high. The data displayed by this command can be reset only when the system is rebooted.
Netstat-s displays the status of each Protocol packet in use. Use the following methods to check the error data: 1. Use ping-f to send a large number of packets to the host. 2. Use netstat-s to check system statistics. 3. Check the value with bad verification in ip, icmp, tcp, and udp. 4. Divide the preceding values by the total number of packets received, that is, the percentage of error packets.
Generally, the error data packet ratio should be less than 0.01%.

Related Articles]

• SCO TCP/IP network management-daemon inetd
  

• SCO TCP/IP network management-daemon)
  

• SCO TCP/IP network management-FTP Configuration