Web Server Outage Diagnostic method

If you are facing Web server outages, the following tips can help you solve the immediate problems and avoid the recurrence of such incidents.

past - years, the Internet has become an important part of civilization. From the social demand level, the network demand ranking is only lower than the drinking water and reliable power supply, the network for human society to provide more productivity, personal entertainment and overall quality of life improvement in the past half a century of development, is no longer comparable to any time.

by visiting the site to meet the requirements has been accepted by many modern users: you can pass the Span style= "font-size:16px" >web for academic research, personal enrichment and pizza ordering, users can do all this in a comfortable home. In order to compete, the most reputable enterprises are also engaged in the world-class web site construction, in general, is to drill down web server development, configuration and maintenance. As businesses increasingly rely on their web Span style= "font-family: Song body; FONT-SIZE:16PX "> Services, these server configurations are also becoming more complex and require more attention to contingency plans to address server or cluster outages. managed server Enterprises and even individuals in a large growth trend, the face of the server should be more aware of the response.

Redundancy and Web Server

before proceeding to the next step, let us emphasize the importance of redundancy. In the past, redundancy solutions were different, but all involved procurement and deployment of physical hardware. Today, the widespread use of virtualization has created a snowball effect, affecting the design and implementation of redundant platforms. Enterprise-class environments still require multiple servers, but not as many as you can imagine. No matter which virtualization platform, just choose one without detailed planning design, the distance to reduce the Web server downtime and failure time objectives, there is a long way to go.

Steps 1 : What's going on? ?

the system administrator must be responsible for analyzing and judging Web responsibility for server downtime reasons. The following issues need to be addressed:

? Whether it is a power outage, generator testing and other similar problems affecting the overall physical environment ?

? and Web is the server's communication completely blocked, or is some IP segment can still be used ?

? Whether you can also manage servers ?

is there an exception record in the log? ?

These are just a few questions that administrators need to answer immediately, and solve the problems before they are thoroughly diagnosed.

Steps 2 : The simplest solution is often the best

there are many times when you get caught up in the pitfalls of diagnosing advanced problems and analyzing various possible technical problems and incurable diseases. In fact, taking a step back, thinking from a macro perspective can solve technical problems that have been spent countless hours and costs that have yet to be overcome. For example, does the host power up, or has someone accidentally knocked off the power cord ? If the machine does already have power, but does not have any network connection, need to check whether the network cable or fiber is not connected or loose. Yes, these seemingly obvious, low-error solutions are unlikely to happen, but any experienced system administrator will tell you that these situations occur much more frequently than you think.

after the two steps above, if your Web server or downtime, don't worry, there are three steps below.

Steps 3 : What to do if the basic fault handling method is not effective

now that you have checked all the cables and other peripherals, you can try to usePingcommand to detect the condition of the device. Fortunately,Pingcommands are common to all platforms and are the simplest. If it can be in the LANPingserver, then try to get out of the LANPingServer for detection. This allows you to quickly determine whether the problem occurs at the interchange and routing levels, not at the server level. In addition, ifWebthe server has been virtualized, trying toPingphysical server's own realityIP. This will help you isolate the problem further. If you are completely unable toPingwith the server, and have determined to fully check the network connection, then more in-depth analysis is needed.

Steps 4 : Make every effort, the server is still down

The network cable has been checked. Also tried to ping the server, still unable to access the server. The good news is that you can already locate the problem to the physical server or the operating system itself. In other words, it is already possible to start a centralized manager to troubleshoot existing problems.

Next, we go from the bottom to the top-level to check the problem, first check the network interface and local network configuration is normal. DHCPwhether to start? Webdoes the server point to the correctDNSServer?If so, you can use the operating system platform to checkWebwhether the service is turned on properly. In theWindowsenvironment, you need to check if the server hasWebthe role of the service. In theLinuxenvironment, the inspection will be more complex, you can try to findhttprelated files or services to ensure that the server is running.

Steps 5 : The desperate need to use a trick

If none of the above works, check the log and try to find out Web The information that is logged in the log when the server is down. Sending this information to more experienced professionals in the area of fault handling and resolution may help. Similarly, if you have already confirmed that the network connection is not a problem, you can use the Wireshark capture tool to assist with the problem by fetching data from the network that is being streamed.

In summary, there are a variety of reasons for server outages. Power outages, misconfiguration, incorrect firewall settings, and even malicious traffic from the Internet can cause source-station outages and system administrators to freak out. All of these issues are sufficient for enterprise decision-makers to pay attention to the design and implementation of redundant solutions, as well as the design and development of the fault-handling process, based on the actual situation of the enterprise's own network.

Telecom server   preferential new and old customers   original price 2400 meta now 1100 meta

Hunan IDC room   High anti-high with Baidu IP segment Dell Brand Server

G Port Access ! Stable! High-protection! Second Solution! Real Defense!

Disregard cc, 7*24*365 hours after-sales operation and maintenance technical support

cooperation small New knock 2851506992 Electricity: 15013023312

Web Server Outage Diagnostic method