Hi,
we are having some random high loads with one of our newly setup servers.
It is built with nginx+apache / php-fpm.
When the load goes high, I get reports similar to one below.
As you can see there is nothing actually that can make the load high. I suspected of the disk, and changed the datastore of this virtual machine, but did not solve the problem.(One option is changing the host machine, but it is quite difficult, I ant to make sure that I eliminate every other possibility) I was once able to log in to machine before it hangs totally once and actually saw some httpd prosesses running, event the report I received from message system was empty like above. Is it possible that php-fpm prosesses are not added to the message system notifications under php-fpm option? Or would systemd proses is the murderer even it shows 0 cpu usage? Thanks to nginx, it gives no server-status page with the free version, we are kind a blind with it. Since my "direct admin" is my wife, I can't purchase a nginx license without her permission.
thanks in advance for any suggestions, ideas.
we are having some random high loads with one of our newly setup servers.
It is built with nginx+apache / php-fpm.
When the load goes high, I get reports similar to one below.
HTML:
One Minute - 50.51
Five Minutes - 58.3
Fifteen Minutes - 43.09
top - 18:30:05 up 1 day, 22:05, 0 users, load average: 50.51, 58.30, 43.09
Tasks: 327 total, 3 running, 324 sleeping, 0 stopped, 0 zombie
%Cpu(s): 23.9 us, 6.2 sy, 10.0 ni, 34.4 id, 24.6 wa, 0.0 hi, 1.0 si, 0.0 st
KiB Mem : 3883128 total, 233544 free, 3261368 used, 388216 buff/cache
KiB Swap: 3905532 total, 1268748 free, 2636784 used. 234092 avail Mem
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
11 root 20 0 0 0 0 S 5.9 0.0 3:54.43 [rcuos/0]
15233 root 20 0 26572 1772 1148 R 5.9 0.0 0:00.02 /usr/bin/top -c -b -n 1
1 root 20 0 209880 1840 848 S 0.0 0.0 0:13.49 /usr/lib/systemd/systemd --switched-root --system --deserialize 24
2 root 20 0 0 0 0 S 0.0 0.0 0:00.54 [kthreadd]
3 root 20 0 0 0 0 S 0.0 0.0 0:49.73 [ksoftirqd/0]
5 root 0 -20 0 0 0 S 0.0 0.0 0:00.00 [kworker/0:0H]
7 root rt 0 0 0 0 S 0.0 0.0 0:00.00 [migration/0]
As you can see there is nothing actually that can make the load high. I suspected of the disk, and changed the datastore of this virtual machine, but did not solve the problem.(One option is changing the host machine, but it is quite difficult, I ant to make sure that I eliminate every other possibility) I was once able to log in to machine before it hangs totally once and actually saw some httpd prosesses running, event the report I received from message system was empty like above. Is it possible that php-fpm prosesses are not added to the message system notifications under php-fpm option? Or would systemd proses is the murderer even it shows 0 cpu usage? Thanks to nginx, it gives no server-status page with the free version, we are kind a blind with it. Since my "direct admin" is my wife, I can't purchase a nginx license without her permission.

thanks in advance for any suggestions, ideas.