Help: High idle CPU usage with High Temperature
-
I hope someone can help me with my issues. By the way, I am a networking noob so bear with me. I did a lot of google search and on this forum but can't figure it out nor saw much help form some similar posts.
Have set up Pfsense box on a used PC.
Pfsense 2.3.2
Dell Optiplex 780 Core 2 Duo E8500 3.16 ghz
WAN port on the port on mobo - 100/1000 (em0)
Lan port on PCI Intel nic card - 100/1000 (em1)
4 Gig DDR3 Ram
HDD is 80 Gig.I have two switches connected (one 8 port TD link, and one 5 port Td link; both are gigabit)
I have a vonage phone box connected to the 8 port switch at default settings.
I have a network printer on the 8 port switchpf box has:
1 VGA connected to a dell monitor / usually turned off.
1 USB keyboard / currently not connected
1 USB mouse / currently not connectedThis is a standard Dell box, nothing fancy.
I have the following packages
Snort
pfblockerNG
darkstat
ifTopWoke up this morning and felt the internet was a bit slow.
**Checked the dashboard and, yikes, I see that the temperatures is at 55C! The two cores are showing 45C and 53C!
And CPU Usage is at 40 to 66%. WTH?**
The normal idle CPU is less than 20% and temperature should be around 40C (37C and 42C; 2 cores).
The memory is around 11%.
There is nothing major running. We don't have any heavy usages in the house.
Can someone please help me?
vmstat -i shows
interrupt total rate
irq14: ata0 331221 1
irq15: ata1 212937 0
irq16: em1 uhci0 15434134485 48327
irq256: hpet0:t0 359912194 1126
irq257: hpet0:t1 79493674 248
irq264: em0 17055007 53
irq265: hdac0 1151 0
Total 15891140669 49758top -s -h gives me
last pid: 13435; load averages: 0.25, 0.47, 0.46 up 3+16:44:46 10:49:43
40 processes: 1 running, 39 sleepingMem: 111M Active, 254M Inact, 312M Wired, 202M Buf, 3106M Free
Swap: 8192M Total, 8192M FreePID USERNAME THR PRI NICE SIZE RES STATE C TIME WCPU COMMAND
23309 root 1 20 0 262M 37560K piperd 0 0:00 0.29% php-fpm
52763 root 3 40 20 613M 223M sbwait 1 0:57 0.00% snort
27047 unbound 2 20 0 67420K 40036K kqread 0 0:40 0.00% unbound
41399 root 1 52 20 17000K 2384K wait 0 0:32 0.00% sh
84412 nobody 1 20 0 16836K 4820K select 0 0:24 0.00% darkstat
20376 root 5 20 0 15012K 2220K accept 1 0:15 0.00% dpinger
20903 root 5 20 0 19108K 2312K accept 0 0:14 0.00% dpinger
24475 root 1 20 0 28896K 7108K kqread 0 0:13 0.00% nginx
80439 root 1 20 0 14508K 2312K select 0 0:13 0.00% syslogd
14188 root 1 20 0 16676K 2728K bpf 1 0:10 0.00% filterlog
24355 root 1 20 0 28896K 7028K kqread 0 0:09 0.00% nginx
46632 root 1 20 0 14408K 1952K select 0 0:08 0.00% powerd
27498 root 1 20 0 30140K 17968K select 1 0:07 0.00% ntpd
268 root 1 20 0 262M 25068K kqread 0 0:04 0.00% php-fpm
15203 dhcpd 1 20 0 20716K 10904K select 1 0:02 0.00% dhcpd
14230 dhcpd 1 20 0 24812K 13704K select 0 0:02 0.00% dhcpd
34577 root 1 20 0 14400K 2240K select 1 0:02 0.00% radvd
24730 root 1 20 0 16532K 2248K nanslp 0 0:01 0.00% cronSystem activities shows below…
Mem: 111M Active, 253M Inact, 312M Wired, 205M Buf, 3106M Free
Swap: 8192M Total, 8192M FreePID USERNAME PRI NICE SIZE RES STATE C TIME WCPU COMMAND
11 root 155 ki31 0K 32K RUN 0 88.2H 100.00% [idle{idle: cpu0}]
11 root 155 ki31 0K 32K RUN 1 82.8H 56.59% [idle{idle: cpu1}]
12 root -92 - 0K 336K CPU1 1 340:20 46.68% [intr{irq16: em1 uhci0}]
69718 root 24 0 262M 32204K piperd 1 0:01 1.27% php-fpm: pool nginx (php-fpm)
0 root -92 - 0K 208K - 0 2:47 0.00% [kernel{em0 que}]
12 root -60 - 0K 336K WAIT 0 1:19 0.00% [intr{swi4: clock}]
0 root -92 - 0K 208K - 1 1:08 0.00% [kernel{em1 taskq}]
5 root -16 - 0K 16K pftm 0 1:06 0.00% [pf purge]I don't have any pollings enabled in networking.
Could it be the network card? It's a generic intel card (it was new) bought off ebay. It worked perfectly when I was using it on a older Dell Athlon box a year ago (scrapped it due to too much electricity usage).
I have no idea what I am looking at or what to do. Would really appreciate any help.
-
Indirectly your network card. Could be a driver or setting issue like you've said, but you started off saying "nothing major is running", but your interrupts are consuming 1/4 of your CPU and 48,327 interrupts per second, which is insane. My Intel i350 only consumes about 600 interrupts/sec(150 per core/IRQ) at 140,000 pps.
-
Indirectly your network card. Could be a driver or setting issue like you've said, but you started off saying "nothing major is running", but your interrupts are consuming 1/4 of your CPU and 48,327 interrupts per second, which is insane. My Intel i350 only consumes about 600 interrupts/sec(150 per core/IRQ) at 140,000 pps.
Thanks for the reply.
Yes, I saw that, too. Really there was nothing running in the morning. What could it be? The traffic was minimal, too, I think. I am mystified. I am a noob at this networking…
Do you think NIC is failing? It's the LAN side intel card on PCI.
-
Indirectly your network card. Could be a driver or setting issue like you've said, but you started off saying "nothing major is running", but your interrupts are consuming 1/4 of your CPU and 48,327 interrupts per second, which is insane. My Intel i350 only consumes about 600 interrupts/sec(150 per core/IRQ) at 140,000 pps.
For the benefits of you and those whom might have interest in this issue, I was able to solve the issue with the high CPU Usage and temperature spikes.
I had the PowerD feature enabled under the System > Advanced > Misc.
The power set ups were all "Hiadaptive" when the box was originally set up. I found out about this somewhere via Google search and that was the clue to change the settings to "Adaptive".
The thread (not on this board) mentioned that it maybe worth to turn off the PowerD feature if the condition does not improve.
I changed the settings to all "Adaptive" and now the CPU usage and temperature is back to normal.
Thanks for the prior reply as well.
-
Ahh, yes. I forgot about PowerD and powerstates for some CPUs/motherboards. Nice catch and thanks for reminding.