Re: [Ntp] Negative leap-second?

On Tue, Aug 2, 2022, 5:19 PM Daniel Franke <dfoxfranke@gmail.com> wrote:

> I think predictions of a negative leap second are very premature, but I
> have fewer worries about a negative one than about a positive one. I've
> reviewed quite a lot of leap second handling code and never encountered any
> that gets the positive case right but the negative case wrong; programmers
> clued into the issue at all and generally clued in enough to know that
> negative ones are a possibility.
>

It's been my experience that even though they know negative leap seconds
are a possibility, they are more apt to have bugs than the positive leap
second code. I've had to fix more bugs in the negative code than the
positive code when we did testing on the GPS simulation system.

What makes me *less* worried about negative leap seconds (rather than just
> not-much-more worried) is their effect on leap-second-naive code: they make
> POSIX time jump forward rather than backward. Lots of code out there
> wrongly assumes that POSIX time can never jump backward. Code that assumes
> it can never jump forward would get broken every time someone every time
> someone suspends their laptop or pauses their VM or just puts their OS
> scheduler under a lot of load.
>

I'd worry about tiny sleeps in the kernel and hardpps() calculations
getting confused as the kernel itself jumps forward a smidge. I wouldn't
rush too fast to say there won't be some subtle problem, likely not in the
time code itself...

The history of leap second code is untested code is buggy code. And
negative leaps are way less tested than positive ones. We've had deadlocks
due to bad lock handling, leap seconds applied at the wrong time (end of
the day when viewed as uptime is a bug I fixed years ago, but that bug was
introduced in the last leap drought around 2000).

So I'd expect it will be basically ok, but less ok than a positive leap...
I would not expect fewer problems...

Warner

On Tue, Aug 2, 2022, 19:06 Danny Mayer <mayer@pdmconsulting.net> wrote:
>
>> Apparently the earth's rotation is speeding up:
>> https://www.timeanddate.com/news/astronomy/shortest-day-2022
>>
>> If we have to subtract a leap-second I think we will see failures, since
>> this is rarely if ever tested.
>>
>> Danny
>>
>> _______________________________________________
>> ntp mailing list
>> ntp@ietf.org
>> https://www.ietf.org/mailman/listinfo/ntp
>>
> _______________________________________________
> ntp mailing list
> ntp@ietf.org
> https://www.ietf.org/mailman/listinfo/ntp
>