On Fri, 5 May 2006, Dave Anderson wrote:
Michael Holzheu wrote:
>
> So I think for very complex installations, there are
> cases, where it makes sense to just snapshot the
> whole system with a crash dump.
>
> But I still think, it is best to just add an elf core command
> to crash and do the rest of user space debugging with gdb...
>
> Michael
>
But like you also mentioned, there's the extra burden of
potentially having to save the swap partition contents
someplace?
That seems kind of ugly...
We actually use the user space function from lcrash quite
frequently at 3PAR. Much of our code lives in user space,
but we try to ensure that critical processes never swap,
so we generally don't mind not having the swap space.
Our situation is pretty specialized, but probably pretty
common with embedded servers...
-castor