The access to the master station is suddenly abnormal

In the afternoon of 20150915, the company's intranet visited the home page of the main station, the quasi-production environment, and the test environment suddenly became as slow as a turtle, and the company's colleagues immediately blew up. Was the website destroyed? many questions sprang up like bamboo shoots.

As an operation and maintenance and network manager, I have the following ideas to solve:

Is there something wrong with the office network?

Log in to the office network router to check, the connection status is ok, there is no large download link for the internal ip, it is normal to open other websites, and the intranet problems are eliminated.

Whether the website has been caused by the concurrency limit.

Log in to the website statistics and monitoring system, there is no large concurrency, log in to the server to view the access log, no abnormal information, the website is excluded.

The code comes online and causes

If there is a problem with the code, the problem will be found immediately after it is launched, just in case, let the developer help query whether there is a problem with the code, and make sure there is nothing wrong with the code (but the problem is in the code, which will be described below)

Firewall policies lead to

If it is shielded by the firewall, the port should be killed directly, which can not be accessed at all and excluded by the firewall.

Ask Ali Yun for help

Since our main station is running on the Ali cloud server, we provide Ali with a work order, hoping to get help, and there are no restrictions on the other side in the end.

After so many attempts above, I did not solve my own depression, ah, one developer said that it was not possible for js to load for a long time, and then load the following content, resulting in the present situation, and the access was rather busy. After such a reminder, I instantly found the problem, and then I found the home configuration file and tested it with the developer. I listed all the third js code that needed to be loaded. One by one, I opened it in the browser and found that there was a js address in it, and it was in the second position. Block this code, and the website would open normally. Ah, ah, ah finally found the problem, so immediately communicated with the third party of js. It turned out that they blocked our intranet ip, finally opened it, and finally solved the problem.

I feel that we must have a clear train of thought to deal with the problem, don't rush, filter slowly.