[PlanetCCRMA] FC14

Kjetil S. Matheussen k.s.matheussen at notam02.no
Sat Jan 1 14:26:10 PST 2011


Fernando Lopez-Lezcano:
>
> [BTW and to be more precise, the current realtime kernel does not allow,
> on purpose, for SCHED_FIFO processes to actually use up _all_ of a
> processor time, if I remember correctly usage is limited to 95% of the
> processor time so that you can recover from errant rt processes - that
> is also never mentioned in the context of security and DOS attacks]
>

This 95% thing does actually have the complete opposite effect.

Earlier, rt_watchdog or das_watcdog almost always saved you from rebooting
the computer when a realtime process ran wild. Now, more usually than
not, the system which is supposed to save you, fails, and you have to
press the reboot button.

But not only that, it also causes both rt_watchdog and das_watchdog
(plus another one, don't remember its name) to fail recognizing an
unresponsive system.

A couple of years ago, I never had to press reboot while developing
realtime software, now it happens quite a lot.



More information about the PlanetCCRMA mailing list