3 in analysing the situation at the end of the server, there is a greater need for extensive knowledge of web applications, hardware performance and configuration of the server, system performance and configuration, network applications and impact on the server。
(iii) tool-type analytical methods
Tool-based analysis methods have powerful testing tools and software that can quickly give analysis of the various parameters of the network or even failures, which are very effective in resolving common network failures。
(iv) a synthesis and empirical analysis methodology built on time, error and successful experience
This approach is applied in the work of most arabic maintenance staff, relying on web-based management and testing tools to quickly locate network failures。
Computer failure to access the internet
In the case of failure of a networked computer, it is first necessary to determine separately whether the computer's card is installed correctly, whether there is a hardware failure, and whether the network configuration is correct in practice by using the return address of the ping machine (127. 0. 0. 1) to determine the correctness of the card hardware installation and the tcp/ip protocol。
If pinghong is able to do so, that would make this part clear. In case of overtime, check whether the computer's card is in conflict with other equipment on the machine. Check the device manager in the system attributes to see if there is a yellow exclamation mark or a red fork in front of the network adapter's equipment, and if there is any indication that the hardware driver was not installed, then it can be deleted and re-installed. In addition, it is necessary to ensure that the tcp/ip protocol is installed correctly and that it is tied to the card you have installed. If ping does not return the address after re-installation, it would be preferable to replace it with a normal web card. Since vlan is divided in the lan, computers in different vlans have different ip addresses, subnet masks and gateways. To match data such as ip addresses set in machine network properties with connected vlans, otherwise there will be network failure。
When it is ensured that the hardware equipment and network of the computer is properly configured, it is then necessary to check whether the double winch line between the computer and the switch, the rj45 port of the switch, or the switch, is in question. At this point, we want the ping internet computer's vlan gateway, or we need to check the above-mentioned points in several steps。
The simplest method is to check the double winch and test it with a cable tester to detect whether it is broken. There was no problem with the double winch, but it had to see if the switch port was broken. A state-of-the-art light is available at each port of the switch to check with other network operators and, if not, to check. The switch's parameter configuration table is also one of the information commonly used by network administrators and is detected as the network user changes, and if the port signal is not on, the port will be damaged and the jumper will be removed from the normal used port for other reasons and identified as a port problem。
2. The failure of a networked set of computers to access the network is due to a failure of the same number of computers to access the network. First, the commonality of these computers is to be found, for example, if they belong to the same vlan or are attached to the same switch, and if they belong to the same vlan and are connected to separate floor switches, check whether there are acl restrictions on the router, whether the vlan is correctly configured on the router and whether the route protocol (such as our osf protocol) is properly configured. If these computers are the same switches, they should go to the machine room to check whether the switch is free of power or whether the switch has a high cpu load rate and a normal link to the upper network equipment。
Usually, all computers connected to a switch are not able to communicate normally with other computers on the internet, which is a typical switch-mortem phenomenon that can be addressed by restarting the switch. If the failure continues after restart, check all the computers connected to the switch, look at each computer that is disconnected, and locate it slowly to a faulty computer。
The failure usually occurs at one of the ports of the switch which becomes very slow and eventually slows down the entire switch or stack. By checking the condition of the switch through the control counter, it was found that the buffer pool of the switch had grown very rapidly to 90 per cent or more. The reason and the solution is that the original connection on this port should first be replaced by other computers to see if it was caused by the network failure of the computer connected to the port, or to reset the wrong port and restart the switch, which may have been damaged on occasion。
Iii. Common methods for locating and removing failures
(i) police performance analysis
Fault positioning is done through online access to alerts and performance information. Our unit uses the seteview network tube to manage the entire unit of network equipment and to observe more frequently the cpu load rate and the flow of lines. When it is reflected that the connection to the network cannot be reached or the speed of the network is very slow, the connection of computers to the switch can be observed via a network pipe, whether there is a break-up, whether the switch has a high cpu load rate and whether the line traffic is high. By observing the port state of the equipment, it is not normal for the computer to be packaged at which port of the analysis and observation exchange。
(ii) access to network equipment logs
Check regularly the logs of network devices and analyse their status. On the basis of my showlonging order, i observed that the 2950 switchboard, which was attached to the 4006 switch, often fell down every 7 hours, and thenup, which, owing to a longer time interval, did not feel the network was disrupted, during which time we checked and determined that the fibre-optic cable, the fluorescent transceiver, the network, the switch configuration and the switch port were normal, and the subsequent interval was reduced from 7 hours to 7 minutes. We therefore immediately determined that the 2950 switch was malfunctioning itself, and immediately replaced the standby switch, which was ready, thereby reducing the time available for trouble management and restoring the network in the shortest possible time。
(iii) replacement method
The replacement method is the replacement of an abnormally functioning object with a normal working object, thereby serving the purpose of locating and defusing a malfunction. The object here may be a cable, a device and a module。
(iv) configure data analysis
Queries, analyses the configuration data of the current equipment and locates malfunctions by analysing the normality of the above configuration data. Reconfiguration is required if there is an error in the configured data。
Concluding remarks
Computer network technology is rapidly evolving and network failure is complex, and this paper presents a number of common types of failure and their maintenance methods. In order to maintain the network in a timely manner in the event of a network failure and to restore the network's normal functioning as quickly as possible, the following aspects of network maintenance also require attention:
Create complete group network documents for query while maintaining them. Such as system needs analysis, overall thinking and programme for network design, planning of network sprawl structures, choice of network equipment and lines, network wiring, network ip distribution, network equipment distribution, etc。
2. Good habits for web maintenance logs, in particular some low-probability but high-risk failures and some high-probability failures, require complete maintenance files for each machine to facilitate subsequent failure clearance. It is also an accumulation of experience。
3. To increase the awareness of cybersecurity, improve the reliability of passwords and add to the mainframe patches and firewalls of up-to-date operating systems, hacking programs, etc. To prevent possible loopholes。
Return directory: a summary of the essays of the web-planning designer
Editor's recommendations: classics for web planning designers 2013 cheng
Mentoring of web-planning designer chapters in 2013 cheng
