using kdb on Red Hat/SGI linux-2.4.18-18SGI_XFS_1.2pre5

Foris, Jim (MED) foris at mr.mr.med.ge.com
Thu Feb 6 15:22:25 PST 2003


Keith Owens wrote:
> On Thu, 06 Feb 2003 16:33:43 -0600, 
> "Foris, Jim (MED)" <foris at mr.mr.med.ge.com> wrote:
> 
>>Second, I am not able to break into a system hang - making the debugger
>>useless for what I really need it for.
> 
> 
> Use the nmi watchdog, boot with nmi_watchdog=1.
> 

Did that... made no difference.

We think it is because CPU0 handles all NMIs.... so if it dies,
so does the NMI watchdog.

> 
>>Is there anything I need to do special to get symbols in the KDB tracebacks
>>(like configure in stack frame pointers, maybe?) ?
> 
> 
> kallsyms is automatically included.  What does your backtrace look like?
>

Here is a section from a crash:


[1]kdb> bt
ESP        EIP        Function (args)
0xc46cde28 0xc01ed880 <unknown>+0xc01ed880
                                kernel <unknown> 0x0 0x0 0x0
0xc46cde2c 0xc46cdde0 <unknown>+0xc46cdde0
                                kernel <unknown> 0x0 0x0 0x0





More information about the kdb mailing list