It can be so (Docker disk performance) that the disk size of the Docker container causes the server disk to be full,

Server exception is another day when I open my beloved server happily; Yeah! Is that right? My server domain name cannot be accessed, and it has been timed out to circle. OK, open ssh to have a remote look, yeah!!! Still can’t access, is it down? Open the Alibaba Cloud control panel to check the server with a feeling of holding stool

Open the monitoring of Alibaba Cloud control panel and find two unusual data

I just hang two websites and one Qinglong panel, shouldn’t it take up resources so much? Remember that when I opened the Qinglong panel two days ago, there was a bit of a card. I left a mind. Wait, wait, the CPU calculation is high. It’s okay. The disk (the first picture, the dark blue below is read, and the top curve is write) has been writing for so long. Is it tired? I have a total of 60 G disks. Now I’m afraid the disk has exploded. Who killed such a dog?.

According to the above information, the disk is full and the server is down. Then I can only restart the server in the control panel, log in with ssh, and use the df command to view the disk information.

I saw it. I saw it. The most used memory is the/directory (root directory) of my system disk. I skillfully took out the ancestral rm – rf/*. No, no, no, no, this is my own server. I can’t use it this way. Forget it, step by step, go to see what happened to the root directory first,

Cd/, let me use du – sh * Kangkang. How is it developing? Look, there is a folder called var that is quite large, 54G. It should be him. Which program in it should have a problem? Let’s take a closer look at what is in it.

To avoid excessive urea, I directly took a screenshot of the problem file, but each level of directory can be checked by du – sh *. Note: the du command checks the file or directory size. The specific parameters can be viewed: the du command describes how to solve the problem. I came to the root of the problem and found that the data under this directory is large.

Let’s see what this path is

Check the content. It is not recommended to use cat, otherwise the server may be too laggy. I will use tail to check the last few lines of tail core.8252

All these things; Seen=not seen; Let me open a smaller file, a file called core.26822, and use cat core.26822 to view the following contents

The path I saw before is that in Docker’s ql, ql is my Qinglong panel, and emmm is one of the things that hang on behalf of me. Then there are some errors in it, which is the keyword. It should be that I have a script running all the time, but the error is reported again. Plus, I didn’t care, which led to its downtime today. I’m sorry that my father didn’t take care of you. Then I went to visit Qinglong panel, which has been jer. OK, then I waved my hand, Play again

Rm – rf core

Well, now my baby is very healthy. Let me start the Qinglong panel to see what’s going on.

Well, it doesn’t look so healthy, so I’ll install a docker for him tomorrow.

Source: Report/feedback

我的科技记录 » It can be so (Docker disk performance) that the disk size of the Docker container causes the server disk to be full,