My Network Card is continuse go down…..

[averykao]

I found some thing ..  cpu loading is high …but ....this firewall have no one on line (just a testing router).....and there is no install any other packegs on this firewall ....

==========================
last pid: 36430;  load averages:  1.59,  1.71,  1.67  up 0+01:35:40    21:32:58
108 processes: 3 running, 87 sleeping, 18 waiting

Mem: 77M Active, 13M Inact, 48M Wired, 224K Cache, 26M Buf, 838M Free
Swap: 2048M Total, 2048M Free

PID USERNAME PRI NICE  SIZE    RES STATE    TIME  WCPU COMMAND
  259 root    131  20  3352K  1176K RUN    65:50 66.06% check_reload_status
33519 root      76  20 82736K 29528K nanslp  0:01 56.79% php

[cmb]

high CPU is a symptom of the link cycling, the root cause is the link cycling. Figure out why you're cycling link, what are you plugged into, is there a cabling issue, NIC issue, something else going on there.

[dotdash]

I had a similar issue with some on-board fxp interfaces. I found manually setting them for 100/full-duplex made them stable. It's worth a shot.

[averykao]

@dotdash:

I had a similar issue with some on-board fxp interfaces. I found manually setting them for 100/full-duplex made them stable. It's worth a shot.

ok  ,  i will try it later..

but now , i turn off webConfigurator  Protocol (https) , to use http at pfsense…
then the cpu loading is down....then the NIC is not going down anymore.....

why ?

[cmb]

It's the opposite, the NIC isn't flapping, so the CPU usage goes down. It has nothing to do with whether you're using HTTP or HTTPS, that's a trivial difference in CPU.

[averykao]

@cmb:

It's the opposite, the NIC isn't flapping, so the CPU usage goes down. It has nothing to do with whether you're using HTTP or HTTPS, that's a trivial difference in CPU.

I really don't know.. but why the process```
and [check_reload_status] is continuse high loading....

======================================
PID USERNAME PRI NICE  SIZE    RES STATE    TIME  WCPU COMMAND
  259 root    131  20  3352K  1176K RUN    65:50 66.06% check_reload_status
33519 root      76  20 82736K 29528K nanslp  0:01 56.79% php

[averykao]

@cmb:

It's the opposite, the NIC isn't flapping, so the CPU usage goes down. It has nothing to do with whether you're using HTTP or HTTPS, that's a trivial difference in CPU.

soory .. cmb

what resion would cause the NIC flapping ?

[cmb]

Figure out why you're cycling link, what are you plugged into, is there a cabling issue, NIC issue, something else going on there?

check_reload_status and PHP have to do many things when link status changes. When it changes non-stop they have to do a lot of things non-stop, using a lot of CPU in the process.

[averykao]

@cmb:

Figure out why you're cycling link, what are you plugged into, is there a cabling issue, NIC issue, something else going on there?

check_reload_status and PHP have to do many things when link status changes. When it changes non-stop they have to do a lot of things non-stop, using a lot of CPU in the process.

thanks a lot for cmb 
I will try to find out the cycling link

[maverick_slo]

This is a driver issue.
Merge all posts from last 2 weeks and you will clearly see that:
a) Cables are OK
b) NICs are OK
c) Switches are OK
d) On some installs like mine there was no change other than pfsense snapshot update…

Regards,
Greg

[averykao]

@maverick_slo:

This is a driver issue.
Merge all posts from last 2 weeks and you will clearly see that:
a) Cables are OK
b) NICs are OK
c) Switches are OK
d) On some installs like mine there was no change other than pfsense snapshot update…

Regards,
Greg

Thanks a lot fot Greg ….

last few days , I check the cables , NIC , Switches . It is really found nothing wrong . And the problem would happen at when I upgrape the kernel.
So , I think , It is driver issue , really ! 
Thanks~

[algowlight]

I had the same problem off and on over this past week. I checked everything I could think of then I called my ISP, we ran a series of test and determined it was a bad NIC. I replaced the NIC and haven't had any problems since. Sometimes it's just a simple hardware problem.

http://www.speedtest.net/result/2687996289.png

[maverick_slo]

Yes, but this is different case…
There are a lot of reports lately about FXP NICs and not all of them have bad hardware...
For bad HW I don`t need to call ISP, I can determine that by myself...