# Intr process on freebsd

**URL:** <https://www.ruby-forum.com/t/intr-process-on-freebsd/209077>\
**Category:** NGINX\
**Created:** [July 5, 2011, 10:39am UTC](https://www.ruby-forum.com/t/intr-process-on-freebsd/209077 "2011-07-05T10:39:04Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![dubstep](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/d/cdc98d/32.png) [@dubstep](https://www.ruby-forum.com/u/dubstep)\
**Post date:** [July 5, 2011, 10:39am UTC](https://www.ruby-forum.com/t/intr-process-on-freebsd/209077/1 "2011-07-05T10:39:04Z")

</div>

Hi  
I am currently investigating moving our platform from linux to freebsd  
On my test server I can see an increase in cpu usage compared to linux  
(specifically system cpu% rather than user%)  
On further investigation, it is the “intr” process using a lot of cpu.  
Nginx is running ~100 req/sec  
Is this normal behaviour and could anyone further describe what’s going  
on/is this normal etc?

12 root 19 -60 - 0K 304K WAIT 0 55:24 7.32% intr

---

<div class="post-metadata">

**Author:** ![rkoskela](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/r/ce73a5/32.png) [@rkoskela](https://www.ruby-forum.com/u/rkoskela)\
**Post date:** [July 5, 2011, 11:05am UTC](https://www.ruby-forum.com/t/intr-process-on-freebsd/209077/2 "2011-07-05T11:05:40Z")

</div>

Hello!

On Tue, Jul 05, 2011 at 08:38:47AM +0000, Richard K. wrote:

> Hi  
> I am currently investigating moving our platform from linux to freebsd  
> On my test server I can see an increase in cpu usage compared to linux  
> (specifically system cpu% rather than user%)  
> On further investigation, it is the “intr” process using a lot of cpu. Nginx is  
> running ~100 req/sec  
> Is this normal behaviour and could anyone further describe what’s going on/is  
> this normal etc?
> 
> 12 root 19 -60 - 0K 304K WAIT 0 55:24 7.32% intr

Try pressing ‘H’ in top to see separate threads, this should give  
you a hint who eats cpu. 8% system at 100r/s seems a bit too many  
for me, too. But I really doubt it’s nginx related.

Maxim D.

---

<div class="post-metadata">

**Author:** ![rkoskela](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/r/ce73a5/32.png) [@rkoskela](https://www.ruby-forum.com/u/rkoskela)\
**Post date:** [July 5, 2011, 11:29am UTC](https://www.ruby-forum.com/t/intr-process-on-freebsd/209077/3 "2011-07-05T11:29:44Z")

</div>

Hmm, it’s the Ethernet  
12 root -68 - 0K 288K WAIT 2 0:37 5.66% {irq256:  
bge0}

What could this mean?

---

<div class="post-metadata">

**Author:** ![rkoskela](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/r/ce73a5/32.png) [@rkoskela](https://www.ruby-forum.com/u/rkoskela)\
**Post date:** [July 5, 2011, 12:47pm UTC](https://www.ruby-forum.com/t/intr-process-on-freebsd/209077/4 "2011-07-05T12:47:39Z")

</div>

Hello!

On Tue, Jul 05, 2011 at 09:29:13AM +0000, Richard K. wrote:

> Hmm, it’s the Ethernet  
> 12 root -68 - 0K 288K WAIT 2 0:37 5.66% {irq256: bge0}
> 
> What could this mean?

With default settings it shows not only ethernet costs but also  
all incoming packet processing costs. You may try setting  
net.isr.direct and net.isr.direct\_force to 0 to see what actually  
takes cpu time, and playing with various network settings  
(including net.isr.\*) to see if it would be possible to reduce  
load in your case.

Maxim D.
