Shulou(Shulou.com)06/01 Report--

At around 9 o'clock this morning, the mobile phone app received an e-mail alarm and opened it to indicate that the site on an ip was abnormal.

Try to connect to ssh and find it very slow. Can't connect properly. It can be connected normally after a while, but the delay is so high that it is basically impossible to operate.

Change the ssh mode and log in through the ssh of other machines in the same intranet and find it normal. PS: in fact, when you are here, you should be able to initially conclude that it is the problem of the external network.

Top view, cpu, memory is normal, netstat check the number of web connections, concurrency about 300. For apache+php+mysql sites, there is no pressure for such concurrency. Io is also normal.

Stop the httpd service and the network returns to normal. PS: there's something wrong with the judgment here. I think it's the httpd. And then wasted a lot of time on apache. And did not find a trace of log-related error records.

After a few seconds of turning on the httpd service, the delay increased, the website access was very slow, and it was impossible to load php at all.

PS: when I was here, the miscarriage of justice was the problem of php. In fact, php loading and reading related data requires a relatively large bandwidth.

The attempt to httpd does not load php, and the delay is normal.

Keep in mind the people involved in the cloud provider ustack here. When they asked ustack to check their network for help, they gave a normal reply from the network.

The next step is to keep trying. From the snapshot to the new host and switch the IP past, it is still an exception.

PS: in fact, the problems of programs and apache,php are basically ruled out at this time. Because a snapshot is a snapshot taken from a normal machine the day before yesterday.

Later, in the case of high website delay, through the intranet curl website, it is found to be in a normal state. Here it is concluded that it must be the problem of the external network IP. Then use the window machine of the intranet to visit the web website again to confirm that all the intranet access is normal.

After the situation has been confirmed, communicate with the personnel of ustack, still reply to us, everything on the network is normal, please check by yourself.

In the last few minutes, the public network IP will be changed and the dns resolution will be modified. After the resolution takes effect, the access will be normal.

Postscript: the positioning problem is too slow, a little doubt their own judgment.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.