[c-nsp] CPUHOG log messages

[Tony Li]

To really see what's going on, you need to both see the source and
sometimes have a debugger sitting there with an unstripped binary
version of the image, disassembling the routines as necessary.  Not
found on CCO.  ;-)

Unfortunately, the traceback is only ever going to give you routine
names, which may or may not (frequently not ;-) do what you might expect
from the name.  So the traceback is usually adequate for bug
identification, but not much else.

Tony


> -----Original Message-----
> From: cisco-nsp-bounces at puck.nether.net 
> [mailto:cisco-nsp-bounces at puck.nether.net] On Behalf Of Rodney Dunn
> Sent: Monday, June 05, 2006 7:55 AM
> To: Yuri Lukin
> Cc: c-nsp
> Subject: Re: [c-nsp] CPUHOG log messages
> 
> Best you can do is the crash decoder on CCO.
> 
> If it doesn't provide anything useful open the case is about
> all you can do.
> 
> Yes...you need access to the symbols file for those images
> which are not available to you.
> 
> Rodney
> 
> On Mon, Jun 05, 2006 at 10:18:08AM -0500, Yuri Lukin wrote:
> > Rodney Dunn wrote ..
> > > I thought they integrated the old crash decoder or something
> > > with output interpreter to do it. I honestly don't know.
> > > 
> > > I'm too geeky and just do it manually. ;)
> > 
> > To do it manually I would imagine one would need to know some of
> > the internals of IOS, correct? The reason I ask is that I run into
> > problems such as this one quite often with customer routers and
> > would like to be able to decode the crash on my own as much as 
> > possible before calling TAC. 
> > 
> 
> _______________________________________________
> cisco-nsp mailing list  cisco-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/
>