cpu core 100% on idle

**DDDamian** · 2012-03-23 01:43

Yeah, his file system is a little messed up, and I think it's not even seeing commoncache as it's in a location that isn't mounting correctly.

DanLongman · 2012-03-23 06:51

Hey so just tonight I noticed my CPU starting to go to 100% while idle.
ssh in and run top.

Looks like its upstart-udev-br at 100% cpu?

Weird thing is just got nyxboard. Reboot system cpu usage at idle is ~10% or less, then i plug in the RF transmittor for the nyxboard, jumps to 100% could this be related?

more:
I can confirm this is because of the nyxboard. I ssh'd in ran top and watched; Low CPU usage, And the second I plugged in the RF adapter the proc upstart-udev-br started and was at 100% CPU Usage.
Any idea what is going on?

now i can kill upstart-udev-br with a signal 9 and everything seems to work normally. with low CPU usage.
But what is this daemon and what is going on?

I will turn on debugging and get a log posted right away.

ericab · 2012-03-23 06:57

Dan;

it doesn't sound related but it would be an interesting find if common.cache had something to do with it.......

maybe you could try to move /storage/.xbmc/addons/script.common.plugin.cache
down one level, and reboot. then try to plug in the RF adapter; while monitoring top of course.
report back !

DanLongman · 2012-03-23 07:08

Sure I can do that, Will report back right away.

for s&g's here is my log before touching common plugin cache.:
22:56:09 is when i plugged in rf

http://pastebin.com/x5sx5Z5A

Yes I think this is not related to your issue.

Yeah same thing happens with common plugin cache moved down a level.
Plug in RF adapter, upstart-udev-br jumps to 100% cpu usage and stays there.

like i said if i kill it everything seems to work fine, and so far I have not seen it respawn.
but what is it, and why is the nyxboard rf adapter making it mad.

Thanks for any help.

DanLongman · 2012-03-23 07:21

more: So figured out its actually upstart-udev-bridge
here is the man page:

Code:

dan@XBMCNIX:~$ man upstart-udev-bridge

upstart-udev-bridge(8)                                  upstart-udev-bridge(8)

NAME

       upstart-udev-bridge - Bridge between Upstart and udev

SYNOPSIS

       upstart-udev-bridge [OPTIONS]...

DESCRIPTION

       upstart-udev-bridge  receives  information  about  kernel  uevents that

       udev(8) has completed and creates init(8) events for them.

       It emits events which match the pattern "S-device-A" where 'S'  is  the

       udev subsystem and 'A' is the udev action.  See udev(7) and for further

       details.

       Assuming /sys is mounted, possible values for subsystem for your system

       are viewable via /sys/class/.

EXAMPLES

       net-device-added

              Event emitted when a network device is added.

so not sure if bug with Nyxboard, or ubuntu bug. see bugs posted w/ google search.
And not sure what will happen leaving it not running. shouldn't be a problem, but pain to have to kill -9 it everytime i reboot.

ericab · 2012-03-23 07:27

you could script it;

stick this in /etc/rc.local:

(sleep 20; kill -9 `pidof upstart-udev-br`)&

mason · 2012-03-23 15:39

I experience the same, finally got my replacement from pulse-eight and start using it again. The adapter is to blame for the 100% udev process, can reproduce it with oneiric but don't know how to fix it yet. (yeah you can kill it, but I like to eliminate the root cause).

DanLongman · 2012-03-23 16:22

(2012-03-23 15:39)mason Wrote: I experience the same, finally got my replacement from pulse-eight and start using it again. The adapter is to blame for the 100% udev process, can reproduce it with oneiric but don't know how to fix it yet. (yeah you can kill it, but I like to eliminate the root cause).

Same.

mason · 2012-03-23 20:38

Seems if found a fix, there a bug with upstart logged. They found it through some laptop battery problem (check out the thread), but it seems to apply to me also.
I installed it and everything seems normal, even readding the receiver looks good which resulted in a reproducable 100% CPU load before.

https://bugs.launchpad.net/ubuntu/+sourc...omments/32

DanLongman · 2012-03-23 20:45

Awesome thanks Mason! I just installed the update, Will have to wait till i get home to test it out and make sure issue is fixed, But I am pretty sure its good now! yay!

ericab Junior Member Posts: 47 Joined: Mar 2012 Reputation: 1	2012-03-23 06:57 Post: #43 Dan; it doesn't sound related but it would be an interesting find if common.cache had something to do with it....... maybe you could try to move /storage/.xbmc/addons/script.common.plugin.cache down one level, and reboot. then try to plug in the RF adapter; while monitoring top of course. report back !
	find quote

ericab Junior Member Posts: 47 Joined: Mar 2012 Reputation: 1	2012-03-23 07:27 Post: #46 you could script it; stick this in /etc/rc.local: (sleep 20; kill -9 `pidof upstart-udev-br`)&
	find quote

mason Senior Member Posts: 266 Joined: Sep 2008 Reputation: 0 Location: Germany	2012-03-23 15:39 Post: #47 I experience the same, finally got my replacement from pulse-eight and start using it again. The adapter is to blame for the 100% udev process, can reproduce it with oneiric but don't know how to fix it yet. (yeah you can kill it, but I like to eliminate the root cause).
	find quote

DanLongman XBMC lover Posts: 49 Joined: Jun 2011 Reputation: 0 Location: Regina, SK, Canada	2012-03-23 16:22 Post: #48 (2012-03-23 15:39)mason Wrote: I experience the same, finally got my replacement from pulse-eight and start using it again. The adapter is to blame for the 100% udev process, can reproduce it with oneiric but don't know how to fix it yet. (yeah you can kill it, but I like to eliminate the root cause). Same. http://www.netresults.ca
	find quote

mason Senior Member Posts: 266 Joined: Sep 2008 Reputation: 0 Location: Germany	2012-03-23 20:38 Post: #49 Seems if found a fix, there a bug with upstart logged. They found it through some laptop battery problem (check out the thread), but it seems to apply to me also. I installed it and everything seems normal, even readding the receiver looks good which resulted in a reproducable 100% CPU load before. https://bugs.launchpad.net/ubuntu/+sourc...omments/32
	find quote

DanLongman XBMC lover Posts: 49 Joined: Jun 2011 Reputation: 0 Location: Regina, SK, Canada	2012-03-23 20:45 Post: #50 Awesome thanks Mason! I just installed the update, Will have to wait till i get home to test it out and make sure issue is fixed, But I am pretty sure its good now! yay! http://www.netresults.ca
	find quote