Why Load Average > 2000 on a normally working server?

Question

linux

Today I discovered a stable increase in load average. Now it's already around 2100. The server is working fine.

top - 05:51:05 up 51 days, 19:19, 2 users, load average: 2152.96, 2105.26, 1978.46
Tasks: 169 total, 1 running, 168 sleeping, 0 stopped, 0 zombie
%Cpu(s): 1.1 us , 0.7 sy, 0.0 ni, 98.1 id, 0.1 wa, 0.0 hi, 0.1 si, 0.0 st
KiB Mem : 2047836 total, 82308 free, 1407496 used, 558032 buff/cache
KiB 398712 availMem

Has anyone experienced this?

Answer 1

Any process can set the "blocked" status, usually when accessing a disk or other IO, the kernel takes this into account when calculating loadAverage, as a result, if some application set this status and left it hanging, then loadAverage will grow.
Look for processes with status D (eg via ps aux), in theory you should have at least 7 such processes.

Answer 2

There are many processes there. Log in, see what's going on.
LA does not always mean that the server has everything)