Re: Locking processes against being killed when low on memory

["Steven M. Bellovin" <smb%cs.columbia.edu@localhost>]

On Fri, 22 Feb 2008 13:38:03 -0500
Thor Lancelot Simon <tls%rek.tjls.com@localhost> wrote:

> On Fri, Feb 22, 2008 at 12:14:48PM +0000, Steven M. Bellovin wrote:
> >
> > I'll try, but with no message and no dump that isn't easy...
> 
> With no message and no dump, are you _certain_ the X server didn't
> just run out of memory and die ungracefully, leaving the system in a
> hopeless state due to its direct frobbing of memory and device
> registers?
> 
I couldn't ping it from the outside, either, and the power button
didn't fire up an ACPI script.  If it was just out of swap, *something*
should have been killed, leaving RAM for the latter.  Nor was X under
any stress, since I all of the other applications active should have
been more or less sitting there.

So -- no, I'm not certain, but to me the evidence is pointing to towards
a crash.


                --Steve Bellovin, http://www.cs.columbia.edu/~smb