Re: [RFC][PATCH] timekeeping: Limit system time to prevent 32-bit time_t overflow

[Miroslav Lichvar]

On Wed, Apr 15, 2015 at 10:31:54PM +0100, One Thousand Gnomes wrote:
> On Wed, 15 Apr 2015 17:41:28 +0200
> Miroslav Lichvar <mlichvar@xxxxxxxxxx> wrote:
> > larger value. When the maximum is reached in normal time accumulation,
> > the clock will be stepped back by one week.
> 
> Which itself is open to exploits and dirty tricks and causes bizarre
> problems.

Any examples? I think it shouldn't be any worse than having system
clock with incorrect time and making a backward step, which is a well
understood problem.

> IMHO it doesn't actually improve the situation.

Do you have a 32-bit system for testing? Try "date -s @2147483600",
wait one minute and see if it's not worth preventing.

I think the power consumption alone is worth it. If there is some
widely used application/service in which the overflow triggers an
infinite loop making requests to a network service, maybe it could
prevent a DDoS attack.

-- 
Miroslav Lichvar
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/