We found we got leap seconds added on some systems over the weekend. There
were no leap seconds planned (
http://www.usno.navy.mil/USNO/earth-orientation/leap-second-announcement),
however some of our systems got one.
We run our own s2/s3/s4 system, with only the s2s going to the Internet.
We have about 20 servers defined there, but looking through the logs, I
can't figure out which one(s) may have been advertising the leap second. I
went through all our systems on Friday and Saturday to check for the leap
bit, but had nothing, so it must have come out on Sunday.
Anyone else run in to this, or have any further intel about servers that
advertised the leap second?
Cheers,
Todd.
We found we got leap seconds added on some systems over the weekend. There
were no leap seconds planned (
http://www.usno.navy.mil/USNO/earth-orientation/leap-second-announcement),
however some of our systems got one.
We run our own s2/s3/s4 system, with only the s2s going to the Internet.
We have about 20 servers defined there, but looking through the logs, I
can't figure out which one(s) may have been advertising the leap second. I
went through all our systems on Friday and Saturday to check for the leap
bit, but had nothing, so it must have come out on Sunday.
Anyone else run in to this, or have any further intel about servers that
advertised the leap second?
Had a leap happen here on the 30th. My stratum 1 source is a CDMA
timekeeper, I'll ping the operator of it and see if he knows anything or if
it logged anything. It's probably not isolated at all since all my S2
machines have some diversity in alternate time sources but still took the
leap second.
We found we got leap seconds added on some systems over the weekend. There
were no leap seconds planned (
http://www.usno.navy.mil/USNO/earth-orientation/leap-second-announcement),
however some of our systems got one.
We run our own s2/s3/s4 system, with only the s2s going to the Internet.
We have about 20 servers defined there, but looking through the logs, I
can't figure out which one(s) may have been advertising the leap second. I
went through all our systems on Friday and Saturday to check for the leap
bit, but had nothing, so it must have come out on Sunday.
Anyone else run in to this, or have any further intel about servers that
advertised the leap second?
We did get an advisory from Infoblox about a bug in "NTP servers based
on open source NTP" that would do just that. For Infoblox NIOS there
was a hotfix, and Symmetricom also has a patch out.
OK he's checked, nothing unusual in logs....data on the box matches NOAA
site (16 leaps, 16 future) - and pool.ntp.org/scores thinks that it's all
OK/well within norms.
My S2s are Symmetricoms, so we may have a winner here.
Cheers!
On Tue 2013-07-02T10:23:58 -0400, Todd S hath writ:
Anyone else run in to this, or have any further intel about servers that
advertised the leap second?
David Malone has been monitoring the NTP pool for years. See his plots
http://www.maths.tcd.ie/~dwmalone/time/leaps/
This time pool was much better than it has been.
I had a quick look at the data, and only 5 of the servers that I
was monitoring advertised a leap on June 30th - three in the US,
one in Argentina and one in New Zealand. If Todd or Michael want,
we can compare notes and see if they are peering with one of the
servers that I spotted.
David.
This might sound like an easy question, but how do you verify if a Red Hat
box took a leap second?
-Grant
As far i can remember the ntp logs have that info... greap for leap pr leap
second
Hi Grant,
My Linux boxes have usually logged a message like:
Jul 1 00:59:59 aturing kernel: [3812251.350269] Clock: inserting leap second 23:59:60 UTC
This message is logged by the kernel, so you can see it in the
output of dmesg - otherwise check in /var/log as it should have
been logged to some one of the files there.
David.