Terminal Server faults and Solutions

As a cost-effective network device, terminal servers have a broad market in China. It can connect many asynchronous devices and only occupy one network node, thus reducing the collision rate on the Ethernet and improving the communication performance of the network. The terminal server can also be used as a printing server or an asynchronous router.

When a problem occurs on the terminal server, it will have a great impact on many users. Next I will introduce two problems and solutions for the JETSTREAM 8500 series terminal servers produced by Specialix in the UK.

Example 1: One day, a colleague on duty called to say that a Terminal Server encountered a problem and the terminal could not be used. I think: the operating voltage of the terminal server is 110 V-240 V. In the past, the terminal server configuration information was lost due to unstable field voltage. Is this the same case? Take A multimeter to the site, the voltage is unstable, the peak voltage reached more than 250 volts. After the terminal server is retrieved, use the default recovery switch to restore all ports to the control port, connect them to the terminal with a serial port, and log on to the terminal server, run the screen command in the command line mode to enter the full screen mode. Configure the system information of the Terminal server under the server configuration menu. Under the line configuration menu, Configure port 1-15 of the terminal server to the application port, and port 16 to the control port. Then, select Save and exit under the command line mode menu. Now, the terminal server configuration is complete. Finally, in order to prevent the occurrence of similar incidents, a UPS is installed while the terminal server is installed on site. Now everything should be done. I did not expect that after the installation, the on-site terminal still cannot be used. What is the problem?

Connect the terminal to the control port 16 of the terminal server. Check the full screen mode and find that the port configuration information of the terminal server is intact, but the system configuration information has changed, the internet address is changed from 172.21.91.4 to 172.21.0.4. Restart the instance after reconfiguration. log on to the instance again. As a last resort, in order not to affect production, we had to replace it with a backup device and take it back for further research. Although I have carefully studied the device instruction manual and relevant technical documents I have accumulated, once the machine is restarted, The internet address is changed to 172.21.0.4 in full screen mode, it was puzzling. After a day, I did not find a solution, so I had to seek help from the factory office in Beijing.

After listening to my introduction, the technical support staff of the manufacturer asked me to use the show server command to view the server configuration in the command line mode, then run the following command in command line mode:

Set server dhcp off

Then configure. After the configuration is completed according to the technical staff, the system returns to normal. Why? Later, I used the show server command in the command line to view the server's system configuration information as follows:

JS_8500 # sh server

Servername ytr4

Internet address 172.21.91.4

Subnet mask limit 255.0

Broadcast address 172.21.91.255

Domain name

Ip_host 0.0.0.0

DNS

Primary 172.21.91.124

Tftp retry 5

Tftp timeout 3

Security off

Authentication both (local + radius)

Dhcp off

Gui access on

Banner off

JS_8500 #

View the server configuration option of the server in full screen mode, as shown below:

Servername [ytr4]

Internet address [172.21.91.4]

Broadcast address [172.21.91.255]

Subnet mask [255.255.255.0]

Domain name []

Ip_host []

Authentication [both (local + radius)]

Gui access [on]

Banner [off]

Comparing the two, we can find that DNS, tftp retry, tftp timeout, and dhcp (Dynamic Host Configuration Protocol) parameters are more displayed in the command line mode than in the full screen mode, because dhcp is in the on status, the terminal server sends a DHCP request to the DHCP server on the Internet. The authorized DHCP server provides an IP address to the server, this IP address is not the host address we need, and thus cannot be used by the terminal. At the same time, because you are used to full-screen configuration, but the existence of DHCP parameters is not displayed in this mode, the cause of the error cannot be found.

Example 2: One day, the on-site staff reported that the display speed of the terminal had slowed down since last night. In the past, when I used this terminal server, the display speed of the work terminal was slow due to failing to comply with the standard wiring. Later, I re-wired the terminal under the guidance of the manufacturer's technical staff, and the terminal work recovered to normal, since then, there have been standard wiring. Why are there problems?

At the scene, the situation is exactly the same as the previous phenomenon. In an office, the display speed of the five terminals directly connected to the terminal server is very slow. In the application, it takes 3 or 4 seconds to press enter to display the menu and use the left and right arrows in the menu bar. We set up optical cables at the scene in the data center, which are then separated by the Hub. One Terminal Server is connected to the Hub, and all terminals are mounted to the terminal server. Five terminals are directly connected to the terminal server using five types of cables, and one terminal is connected to the terminal server using a baseband digital transmitter over a leased line. From the perspective of the phenomenon, there is no possibility of a problem with the terminal device. Use the slave device to check the normal Terminal Server elsewhere to replace the device here. The phenomenon remains the same. In this way, the line is left when the problem occurs. First, we first find a computer connected to the on-site Hub, and use the terminal simulation program to experiment. The result shows that the optical cable is normal. Then I used a new five-class line to connect terminal servers and terminals. The display speed of the terminal was very slow. It seems that the original terminal line is no problem. Where is the problem?

I accidentally heard that there was a problem on the station and the signal was disturbed. Will it be electromagnetic interference? As a result, the user who connects to the terminal server through the baseband data transmission is normal, and the remote user is not affected. Now that the problem is found, the terminal server is moved to the centralized Office of the terminal, and the remote user line is changed to the office. After the connection is completed, the application is run and the normal speed is finally restored. Later, it was confirmed by the power department that the high-voltage lines near the site had been transformed. Post-event analysis: the network devices used in the network are 10 M. Although electromagnetic interference exists, the impact on the microcomputer is minimal. The connection speed between the terminal and the terminal server is only BPS, and the interference effect is naturally obvious. Remote users use a base-band digital transmitter to zoom in the transmission signal. Of course, they can avoid interference.

We can see from the above two examples that we must fully consider how to handle the problem. In the first example, if you do not only work in full screen mode, you can solve the problem by turning to the command line method for search. In the second example, if you only consider the problem of the Microcomputer equipment, without thinking about the impact of the external environment, the problem still cannot be solved. Only by emancipating the mind can we quickly find the problem and solve the problem, so that the network can be unobstructed.

1. How to set the SQL Server replication structure between the three servers
2. Implement effective management of multiple databases on one server
3. Avocent releases software management service package