SPLAT NGX and high CPU utilisation

[glc73]

Hello,

do you have any idea why my SPLAT NGX, when doing nothing, an "uptime" says :

10:23:07 up 48 days, 22:30, 1 user, load average: 3.00, 3.00, 3.00

Is there any errors in the CPU load ? locked on "3.00" while doing nothing is quite surprising....

[kva.kva]

What do command "top" display?

[glc73]

here is a snapshot of "top" output :

10:40:00 up 48 days, 22:47, 1 user, load average: 3.01, 3.01, 3.00
50 processes: 49 sleeping, 1 running, 0 zombie, 0 stopped
CPU states: cpu user nice system irq softirq iowait idle
total 0.4% 0.0% 1.0% 0.0% 4.8% 3.4% 190.2%
cpu00 0.4% 0.0% 0.6% 0.0% 2.6% 1.6% 94.8%
cpu01 0.0% 0.0% 0.4% 0.0% 2.2% 1.8% 95.5%
Mem: 2055260k av, 1108216k used, 947044k free, 0k shrd, 100152k buff
267992k active, 572104k inactive
Swap: 4194224k av, 0k used, 4194224k free 837204k cached

PID USER PRI NI SIZE RSS SHARE STAT %CPU %MEM TIME CPU COMMAND
13126 root 15 0 23660 23M 9588 S 0.9 1.1 0:24 0 fw
5666 root 15 0 2340 2340 1940 S 0.1 0.1 0:00 0 sshd
1 root 15 0 504 504 448 S 0.0 0.0 0:28 1 init
2 root RT 0 0 0 0 SW 0.0 0.0 0:00 0 migration/0
3 root RT 0 0 0 0 SW 0.0 0.0 0:00 1 migration/1
4 root 15 0 0 0 0 SW 0.0 0.0 0:00 1 keventd
5 root 35 19 0 0 0 SWN 0.0 0.0 0:00 0 ksoftirqd/0
6 root 34 19 0 0 0 SWN 0.0 0.0 0:00 1 ksoftirqd/1
9 root 25 0 0 0 0 SW 0.0 0.0 0:00 1 bdflush
7 root 15 0 0 0 0 SW 0.0 0.0 0:12 1 kswapd

[kva.kva]

Do you have SmartPortal on your server or enabled WebUI?

And do you have SmartView Monitor? May be you find in it interesting information?

And sort "top" by memory usage (M key in "top"'s window).

[glc73]

here is the "top" sorted by Memory usage :

11:03:46 up 48 days, 23:11, 3 users, load average: 3.00, 3.00, 3.00
62 processes: 61 sleeping, 1 running, 0 zombie, 0 stopped
CPU states: cpu user nice system irq softirq iowait idle
total 0.4% 0.0% 1.0% 0.4% 6.4% 5.2% 186.4%
cpu00 0.0% 0.0% 0.2% 0.4% 3.6% 2.6% 93.1%
cpu01 0.4% 0.0% 0.8% 0.0% 2.8% 2.6% 93.3%
Mem: 2055260k av, 1157444k used, 897816k free, 0k shrd, 100284k buff
277432k active, 602784k inactive
Swap: 4194224k av, 0k used, 4194224k free 879372k cached

PID USER PRI NI SIZE RSS SHARE STAT %CPU %MEM TIME CPU COMMAND
13129 root 15 0 49464 48M 14356 S 0.0 2.4 0:05 1 fwm
13126 root 15 0 24292 23M 9628 S 0.7 1.1 0:30 1 fw
13039 root 17 0 15304 14M 9480 S 0.1 0.7 0:26 0 cpd
13148 root 15 0 11968 11M 5804 S 0.0 0.5 0:00 0 cpstat_monitor
13140 root 15 0 11224 10M 5156 S 0.0 0.5 0:00 1 stormc
1192 root 25 0 7308 7300 5392 D 0.0 0.3 0:00 1 vpn
1193 root 25 0 7308 7300 5392 D 0.0 0.3 0:00 1 vpn
13130 root 15 0 6508 6508 5164 S 0.0 0.3 0:00 1 cpca
13025 root 15 0 4480 4480 3276 S 0.0 0.2 0:01 1 cpwd
982 root 25 0 4476 4476 3500 S 0.0 0.2 0:00 1 cpwmd
13245 root 15 0 4224 4224 3224 S 0.0 0.2 0:00 0 cprid
986 nobody 25 0 4164 4164 3228 S 0.0 0.2 0:00 0 cp_http_server


In the Smartview Monitor, it says that CPU usage is about 0% or 1%...
really strange, my disks are SCSI in raid1...

[kva.kva]

http://www.redhat.com/archives/rhn-u.../msg00061.html
"A load average is literally the number of processes "running" during any given sample interval. The Linux kernel determines this load every 10 seconds or so, and averages this over a minute (so six samples). That becomes your one minute load average.
It has *absolutely* nothing *whatsoever* to do with CPU utilization. If you have 100 processes blocking on IO, it is fully possible to have a load of 100 and a CPU utilization of nearly 0%."

I found some forums where this problem is discussed, but nothing about why it happens.
May be problem with "slow" hard...

[omahrez]

Hi,
I have the same problem, when I disable the cluster membership of the standby firewall, the cpu goes to 100% idle.
I am sure that this issue is related to the clusterxl but I still don't know how to solve the problem.... still working on it.

omahrez.