[ntp:questions] Cannot down load the leap file

Martin Burnicki martin.burnicki at meinberg.de
Fri Jan 3 14:50:32 UTC 2014


unruh wrote:
> On 2014-01-02, David Taylor <david-taylor at blueyonder.co.uk.invalid> wrote:
>> On 02/01/2014 04:42, Williams Catherine wrote:
>>> Dear,
>>>
>>> I'm not sure why I cannot download the leap seconds file.  Where can be
>>> used now?
>>>
>>>      - ftp://time.nist.gov/pub/leap-seconds.3535142400
>>>      - ftp://tycho.usno.navy.mil/pub/ntp/leap-seconds.3535142400
>>>
>>> BR,
>>> Yang Hong
>>
>> Catherine,
>>
>> These servers appear to be very unreliable, at least when accessing from
>> outside the USA.  There is an alternative at:
>>
>>     ftp://utcnist.colorado.edu/pub/
>>
>> but its file:
>>
>>     leap-seconds.3535228800
>
> No. That file expires Jun 29 2014 which is what you would expect for Jan
> 2 2014. It is not stale (at least assuming it is the same file as that
> same named file on ftp://time.nist.gov/pub/leap-seconds.3535228800
> .
>>
>> is from last March, and is now stale.  The other two sites haven't come
>> up in spite of waiting several minutes (at 08:00 UTC, Jan 02).
>
> I also had trouble with usno, but not with nist.
>
>>
>> For such an important file, I expect there are mirrors in Asia and
>> Europe - it's just that I haven't found them as yet.
>>

Bulletin C46 is from 8 July 2013, and expires 28 June 2014, a few days 
before the next possible leap second date, which would be 30 June 2014.

Usually the next bulletin C is always published a few days after a 
previous possible leap second date. 31 December 2013 was the last recent 
possible leap second date (no leap second has been scheduled, though), 
so I'd expect the bulletin C47 to be published in the next days.

This should also result in an updated leap second file, which should 
expire some time end of December 2014.


In the past I've also found that some web sites which were expected to 
provide a leap second file didn't have the file, or only 
outdated/obsolete version(s) of the file.

IMO best practice would be if the IERS which determines if a leap second 
is scheduled or not, and sends bulletin C to publish the decision, would 
also act as the main authority providing a current master leap second 
file in a secure way, e.g. via SFTP.


Martin
-- 
Martin Burnicki

Meinberg Funkuhren
Bad Pyrmont
Germany



More information about the questions mailing list