Load Average above 20 for single core vps on debian [closed]

Solution 1:

Your server is spending an inordinate amount of time in I/O wait.

57.0%wa

This means... disk.

A likely cause of this problem is the server your VPS runs on is having issues with the disk(s). Those issues include, but are not limited to: a failing disk; using non-enterprise-grade disks; your host trying to run a VPS business on creatively recycled hardware, etc.

It could also be that you are running a process that's causing unusually high amounts of disk activity. Unfortunately that information isn't reported in your top or ps listing, making me suspect you have a low-end OpenVZ based VPS. (Which, if true, puts you back at the previous paragraph.)

As for resolving the problem, the first thing to do is to rule out any of your processes as causing high amounts of disk activity. The iotop program, as mentioned by @Shi, is good for this. Though my bet is you'll find nothing. Once done, you then contact the host to report some issue with the server's disk that they will have to diagnose, since they're the host and you can't see that from within the container.

(And later, when you're shopping for a new VPS provider, steer clear of any who use OpenVZ. It's been my experience that the vast majority of them are run pretty badly.)

Solution 2:

First, try to understand the line below:

Cpu(s): 6.3%us, 14.7%sy, 0.0%ni, 17.5%id, 57.0%wa, 0.1%hi, 4.4%si, 0.0%st

You see that the CPU for user space applications (%us) does nearly nothing. It is busy with system (kernel) tasks (%sy) and even more with waiting (%wa).

Waiting usually means I/O (input/output) waiting. So, check the input/output stats using iotop.

As swap is disabled, swap is not causing this.