[asterisk-users] Asterisk 1.62.13 - CPU spikes every 10 minutes - NOT FIXED

[Mike]

Thanks.  The problem is I only have two PCIe slots, and they are both taken.
I`ll definitely try disabling USB, I have no need for it right now.

 

Mike

 

From: asterisk-users-bounces at lists.digium.com
[mailto:asterisk-users-bounces at lists.digium.com] On Behalf Of Luis Antonio
Prata Barbosa
Sent: Tuesday, September 28, 2010 17:10
To: Asterisk Users Mailing List - Non-Commercial Discussion
Subject: Re: [asterisk-users] Asterisk 1.62.13 - CPU spikes every 10 minutes
- NOT FIXED

 

Hi Mike, 

I`m not sure, but I suggest the following tests.
The cards generate interrupts even with no load. And in real life they have
different clocks. A very small clock difference could accumulate and make
the interrupts happen at same time. USB , TCE400 or PRI card interrupt
routines could be having problems at this moment.

1) Separe usb interrupts from digium boards, try to do it by changing board
slots. (or disabling usb in BIOS).

2) Try to run your system using one digium board at once to determine which
one is generating the problem or if the problem occurs only with both
present.

 
Luis A P Barbosa

2010/9/28 Mike <list at net-wall.com>

(sorry for the series of email, I realize I seem to be having a discussion
with myself).

I was wrong, the "problem" isn't fixed.  Is having IRQ spikes every 10
minutes (under no load at all) the norm with Dahdi hardware?

Mike



> -----Original Message-----
> From: Mike [mailto:list at net-wall.com]
> Sent: Tuesday, September 28, 2010 3:34
> To: 'Asterisk Users Mailing List - Non-Commercial Discussion'
> Subject: RE: [asterisk-users] Asterisk 1.62.13 - CPU spikes every 10
> minutes - FIXED...?
>
> ...which, after upgrading to Dahdi 2.4.0 seem to be gone.  Not sure if
> 2.3.0.1 was the culprit, or just the fact that sonething was stuck
> (stopping dahdi took about 5 attempts because it was apparently in use,
> even if Asterisk had been shutdown).
>
> Hopefully this isn't just the calm before the storm and Dahdi will behave
> under load.
>
> Thanks for the help Shaun, I appreciate you helping point me in the right
> direction. Hopefully this will help someone.
>
> Mike
>
> > -----Original Message-----
> > From: Mike [mailto:list at net-wall.com]
> > Sent: Tuesday, September 28, 2010 2:55
> > To: 'Shaun Ruffell'
> > Subject: RE: [asterisk-users] Asterisk 1.62.13 - CPU spikes every 10
> > minutes
> >
> > Actually, after further investigation I found that every 10 minutes I
get
> > massive irq usage spikes.  The CPU is busy servicing hardware
interrupts.
> >
> > I do have a 4 port PRI card and a TCE400B card,  but tha happens even
> when
> > asterisk activity=0.
> >
> > Mike
> >
> > > -----Original Message-----
> > > From: Shaun Ruffell [mailto:sruffell at digium.com]
> > > Sent: Monday, September 27, 2010 15:58
> > > To: Mike
> > > Subject: Re: [asterisk-users] Asterisk 1.62.13 - CPU spikes every 10
> > > minutes
> > >
> > > When the asterisk process spikes, do you know if there is a
> > > corresponding spike in us, sy, ni, hi, or si time?
> > >
> > > On 09/27/2010 02:53 PM, Mike wrote:
> > > > That`s exactly what I did, and exactly what didn't happen.
> > > >
> > > > Mike
> > > >
> > > >> -----Original Message-----
> > > >> From: Shaun Ruffell [mailto:sruffell at digium.com]
> > > >> Sent: Monday, September 27, 2010 15:48
> > > >> To: Mike
> > > >> Subject: Re: [asterisk-users] Asterisk 1.62.13 - CPU spikes every
10
> > > >> minutes
> > > >>
> > > >> Just something like "echo 1200 >
> > > >> /proc/sys/net/ipv4/route/secret_interval" should have been
> sufficient
> > to
> > > >> half the frequency of the CPU spikes.
> > > >>
> > > >>
> > > >> On 09/27/2010 02:42 PM, Mike wrote:
> > > >>> I just changed the file, did I need to reboot or do anything for
> the
> > > >> change
> > > >>> to take effect?
> > > >>>
> > > >>> Mike
> > > >>>
> > > >>>> -----Original Message-----
> > > >>>> From: Shaun Ruffell [mailto:sruffell at digium.com]
> > > >>>> Sent: Monday, September 27, 2010 15:34
> > > >>>> To: Mike
> > > >>>> Subject: Re: [asterisk-users] Asterisk 1.62.13 - CPU spikes every
> 10
> > > >>>> minutes
> > > >>>>
> > > >>>> Hmm..ok, then I'm not sure I'll be able to help.  It's just that
I
> > > know
> > > >>>> I've seen issues similar to what you're reporting based on the
> route
> > > >>>> table flushing pushing some si time into the asterisk process
(and
> > it
> > > >>>> runs at 10 minute intervals by default).
> > > >>>>
> > > >>>> So I'm not as familiar with the Asterisk code base to know what
> sort
> > > of
> > > >>>> periodic tasks may have changed between the versions you
reported.
> > > >>>>
> > > >>>> On 09/27/2010 02:23 PM, Mike wrote:
> > > >>>>> That wasn't it.  I wonder what happens precisely every 10
minutes
> > in
> > > >>>>> Asterisk (or that is accounted to Asterisk's CPU usage).
> > > >>>>>
> > > >>>>
> > > >>>>
> > > >>>> --
> > > >>>> Shaun Ruffell
> > > >>>> Digium, Inc. | Linux Kernel Developer
> > > >>>> 445 Jan Davis Drive NW - Huntsville, AL 35806 - USA
> > > >>>> Check us out at: www.digium.com & www.asterisk.org
> > > >>>
> > > >>
> > > >>
> > > >>
> > > >> --
> > > >> Shaun Ruffell
> > > >> Digium, Inc. | Linux Kernel Developer
> > > >> 445 Jan Davis Drive NW - Huntsville, AL 35806 - USA
> > > >> Check us out at: www.digium.com & www.asterisk.org
> > > >
> > >
> > >
> > > --
> > > Shaun Ruffell
> > > Digium, Inc. | Linux Kernel Developer
> > > 445 Jan Davis Drive NW - Huntsville, AL 35806 - USA
> > > Check us out at: www.digium.com & www.asterisk.org


--
_____________________________________________________________________
-- Bandwidth and Colocation Provided by http://www.api-digital.com --
New to Asterisk? Join us for a live introductory webinar every Thurs:
              http://www.asterisk.org/hello

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
  http://lists.digium.com/mailman/listinfo/asterisk-users

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.digium.com/pipermail/asterisk-users/attachments/20100928/c00177a6/attachment.htm