Rob,
You are way more polite than I might be.
I do believe, however, that NTF's General Timestamp API project might
help POSIX out here, in that it can map other timescales to UTC,
although going the other direction means one has to know how the POSIX
system handles leap seconds.
POSIX needs at
Martin,
Cosine smearing might need to be a choice. It's harder to track the
leap second if you get a sample during when both phase and frequency are
changing.
--
Harlan Stenn <st...@ntp.org>
http://networktimefoundation.org - be a
many folks are being slow to
update. Dealing with leap second additions and deletions will be yet
another incentive to upgrade this software.
--
Harlan Stenn st...@ntp.org
http://networktimefoundation.org - be a member!
___
LEAPSECS mailing list
library implementation for user-level stuff, and I
expect we'll have GTSAPI implementations built-in to both a linux kernel
and a BSD kernel.
It will take additional resources that we do not yet have to make these
things happen.
--
Harlan Stenn st...@ntp.org
http://networktimefoundation.org
Poul-Henning Kamp writes:
In message 20150426160837.ga11...@ucolick.org, Steve Allen writes:
Both messages show the conceptual failure that results from
Recommendation 460, that is a complete lack of concern for the
distinction between UTC and GMT.
You're so cute Steve...
way you are up for. One of the issues that will need more attention
is pre-1972 stuff.
--
Harlan Stenn st...@ntp.org
http://networktimefoundation.org - be a member!
___
LEAPSECS mailing list
LEAPSECS@leapsecond.com
https://pairlist6.pair.net/mailman
satellite systems use different
timescales then that's perfectly OK too.
As long as there's a way to map these to/from a canonical form
(currently TAI) then we should be fine.
--
Harlan Stenn st...@ntp.org
http://networktimefoundation.org - be a member
Joseph Gwinn writes:
On Sun, 01 Mar 2015 20:35:20 +, Harlan Stenn wrote:
Once people get a system to work, they don't tend to go fixing things
that ain't broke.
There's breakage they know about and breakage they don't know about...
2. Slide titled Possibilities for Future Improvements
/detail.cfm?id=2717320 for more
information.
--
Harlan Stenn st...@ntp.org
http://networktimefoundation.org - be a member!
___
LEAPSECS mailing list
LEAPSECS@leapsecond.com
https://pairlist6.pair.net/mailman/listinfo/leapsecs
for what should be
done when the time steps - those sort of events could be reason to
re-evaluate a significant class of timer events, which includes the need
to re-evaluate trust certificates, which may cause a reload of DNS and
other prior vetted information.
--
Harlan Stenn st...@ntp.org
http
for
an updated file and download it if needed.
If somebody was to do this under a suitable license, I'll happily add it
to the NTP distribution.
This could be done in addition to any other leapsecond distribution
approaches.
Just an idea...
--
Harlan Stenn st...@ntp.org
http
Magnus Danielson writes:
DNSSEC should be part of a DNS solution for instance.
... and correct time is a requirement for DNSSEC.
H
___
LEAPSECS mailing list
LEAPSECS@leapsecond.com
https://pairlist6.pair.net/mailman/listinfo/leapsecs
, but the false leap flags
seem to be way down by now.
This is almost certainly not a problem with NTP. It is almost certainly
a problem with either a refclock (or a signal being fed to a refclock)
or folks having bad leapsecond data, or manually forcing a leap second
notice.
--
Harlan Stenn st
with a new name for the thing they think will solve their problems.
--
Harlan Stenn st...@ntp.org
http://networktimefoundation.org - be a member!
___
LEAPSECS mailing list
LEAPSECS@leapsecond.com
https://pairlist6.pair.net/mailman/listinfo/leapsecs
Clive D.W. Feather writes:
Harlan Stenn said:
I'm still thinking the answer is leave existing 'names' alone - if
you want TAI use TAI. If you want UTC, use UTC. If you want
something new, call it something new.
If people are using a defined name for a defined purpose and it works
for them
Warner Losh writes:
On Feb 12, 2014, at 5:36 AM, Greg Hennessy wrote:
Um, that is false. All linux kernels did not crash, in fact NONE of
mine did.
all here was an overstatement, but the impact of the leap second
should never be your kernel crashes even if your personal kernels
Warner Losh writes:
On Feb 12, 2014, at 7:53 AM, Harlan Stenn wrote:
Warner Losh writes:
On Feb 12, 2014, at 5:36 AM, Greg Hennessy wrote:
Um, that is false. All linux kernels did not crash, in fact NONE of
mine did.
all here was an overstatement, but the impact of the leap
Warner Losh writes:
On Feb 12, 2014, at 1:50 PM, Harlan Stenn wrote:
The conclusions I draw from the utter lack of any similar reports from
non-linux systems are:
- either those kernels/libraries did not do leap-second processing, or
- they did and their code worked
Do you have
Warner Losh writes:
...
A TAI realization of time_t isn't POSIX, which specifically proscribes
UTC.
I think you mean prescribes.
H
___
LEAPSECS mailing list
LEAPSECS@leapsecond.com
http://six.pairlist.net/mailman/listinfo/leapsecs
I just landed, and my sleep clock is seriously disrupted.
Rob, the presentation you have from me is likely the one that was
converted to powerpoint, and it didn't convert as well as I expected.
I'll send a PDF version as soon as I can.
--
Harlan Stenn st...@ntp.org
http
(Found unsent in my Drafts folder...)
Warner Losh writes:
I think the real reason that UT1 shouldn't be considered a time scale
is that it is based on not an imperfect realization of a fixed length
second, but rather an imperfect realization of a variable (measured by
oscillations of a fixed
So I gotta ask.
What's the problem with doing radar and other similar things in GPS time
and keeping human time in UTC, with leap seconds?
I mean, sure, years ago timestamps were YYMMDDHHMMSS and those
eventually got bigger, and eventually folks started noticing that things
really got
it
at the right time, use a different timescale, ...) but it boils down to
making an evaluation and a decision, then implementing and testing it.
--
Harlan Stenn st...@ntp.org
http://networktimefoundation.org - be a member!
___
LEAPSECS mailing list
ntpd can easily track SI seconds or angle time seconds. The
differences are small enough over a day to be easily amortized.
It would not be all that difficult to create an NTPv5 protocol (for
example) that would include timescale as a parameter. There may be a
way to do this using the v4
24 matches
Mail list logo
