Website stability Solution

As we all know, the first major thing to ensure after a website goes online is stability, including the Stability of server space and domain name resolution. However, in some cases, server downtime is inevitable, such as IP address line switching, unexpected power outages, server hardware faults, and hacker attacks, various Causes may cause server downtime. So the current server downtime has become the biggest headache for webmasters, and it is the lingering haze of webmasters. Cluster servers are undoubtedly the first choice to cope with server downtime, but cluster servers generally have high requirements on server hardware, especially the primary server. Although enterprise websites have high requirements on stability, however, the investment is too large and not practical. By taking advantage of the existing conditions, we can develop a stable solution that meets the requirements of enterprise websites, wang Xiaohan will share with you the server downtime solution that Zhengzhou website construction company bayou has already implemented.

1. DNSPOD downtime monitoring

DNSPOD now provides the downtime monitoring and control function, which can be enjoyed by free users and paying users. During domain name resolution, the backup IP address can be resolved at the same time. When the system detects a crash, the IP address can be automatically resolved to the backup IP address. This ensures that website operations are not affected. The actual application of the author finds that the free user's downtime detection is 15 minutes. If the detection frequency limit and TLT are added, the automatic response time after downtime is about one hour, while the VIP user can increase the response time to about 10 minutes. Currently, this function has been used by some customers for Zhengzhou website construction, and we believe it will further ensure the stability of the customer's website.

2. Monitoring treasure url callback

Cloudmonitor provides the url callback function. Currently, this function is only available to VIP users. You can customize the specified callback content based on the callback parameter api documentation. Relatively speaking, dnspod downtime monitoring only monitors the Server IP address, many cases, such as iis failure, apache failure, and website program failure, cannot be detected, but the website has been suspended. The url callback of the monitoring site can solve this problem well. As long as the website cannot be accessed, the monitoring site will immediately notice that if the url callback is combined with the dnspod api documentation, so you can easily develop a custom IP address replacement function. CEN has adopted a dedicated url callback program for department customers to implement server downtime monitoring and switching measures.

3. the perfect combination of chinacache and dnspod

The functions of chinacache are similar to those of cloudmonitor. It also monitors the website running status, but does not have the url callback function. Instead, chinacache has developed APIs for dnspod and dnspod, can directly enter the dnspod account password in the Back Station, automatically complete the url callback control to modify the domain name api function, Lijiang Ancient City Network (http://www.0888123.com) on the function of Wei Cai trial, basically, you can complete the Failover function. If you do not have the relevant technical capabilities, you can try out the monitoring function of the SLB. Unfortunately, the SLB is still not well-developed in many places.

Conclusion

Based on 99.9% of the website's annual operational stability, the allowable downtime rate is about 87 hours, that is, about 3 days. The servers used by general network companies are not very high-end configurations. They host enterprise websites in local IDCs, and combine line faults, hardware faults, and security faults, the annual downtime rate is not small, and the total time should be much greater than three days. This solution is used for configuration and the website resolution will be transferred to the backup server as soon as the downtime occurs, reducing downtime to a minimum is a great benefit to the enterprise's later development.