[ntp:questions] NTP sync failing with error: Leap not in sync

Unknown dallasg79 at gmail.com
Wed Apr 1 20:23:39 UTC 2015

Hi all,
I'm observing some very bizarre behavior from my NTP server.  It's a
fairly straight-forward setup which I found from a "Server World" how-to
article and the system starts up just fine and my clients are able to
sync to the system, no problem.  However, after about 30 minutes or so
of around 60 nodes hitting it, I notice that the clients are unable to
sync against the server.  Here are the troubleshooting steps I've
completed so far:

After observing the issue, I checked the status with ntpdate -d option:

[user at ntp-client ]$ ntpdate -d ntp-server
 1 Apr 14:28:05 ntpdate[18328]: ntpdate 4.2.6p5 at 1.2349-o Sat Dec 20
02:53:40 UTC 2014 (1)
Looking for host ntp-server and service ntp
host found : ntpserver.domain.local
receive( Server dropped: Leap not in sync
server, port 123
stratum 3, precision -23, leap 11, trust 000
refid [], delay 0.02570, dispersion 0.00000
transmitted 4, in filter 4
reference time:    d8c6c506.84fb0c1f  Wed, Apr  1 2015 14:27:02.519
originate timestamp: d8c6c543.8770608e  Wed, Apr  1 2015 14:28:03.529
transmit timestamp:  d8c6c545.2e5eab0a  Wed, Apr  1 2015 14:28:05.181
filter delay:  0.02586  0.02571  0.02570  0.02570 
         0.00000  0.00000  0.00000  0.00000 
filter offset: -1.65207 -1.65213 -1.65213 -1.65213
         0.000000 0.000000 0.000000 0.000000
delay 0.02570, dispersion 0.00000
offset -1.652136

 1 Apr 14:28:05 ntpdate[18328]: no server suitable for synchronization

I searched around on Google and found a suggestion to use the -b option
in the ntpdate command on the server to sync with the external sources
(redhat pool):

[user at ntp-server ~]# ntpdate -u 0.centos.pool.ntp.org
 1 Apr 15:04:41 ntpdate[4207]: adjust time server offset
0.114860 sec

...and to confirm that things looked good on the server:

[user at ntp-server ~]# ntpq -pd
1 packets reassembled into response
     remote           refid      st t when poll reach   delay   offset
2 packets reassembled into response
+ns20.alltraders    2 u   53   64  377   54.031   74.881
2 packets reassembled into response
*cheezum.mattnor    2 u   57   64  377    2.780   74.319
2 packets reassembled into response
+clock.team-cymr      2 u   47   64  377   35.361   97.601
2 packets reassembled into response
 LOCAL(0)        .LOCL.          10 l    -   64    0    0.000    0.000

After updating the clock manually on the server, things seem to run fine
for about 30 minutes or so, when the initial problem presents itself
again...  I know it isn't a connectivity issue because the state goes
from working to not working over a period of time.

[[Configuration Files]]
Here's the sanitized conf file of the server:

driftfile /var/lib/ntp/drift
restrict default nomodify notrap nopeer noquery
restrict ::1
restrict mask nomodify notrap
server 0.centos.pool.ntp.org iburst
server 1.centos.pool.ntp.org iburst
server 2.centos.pool.ntp.org iburst
includefile /etc/ntp/crypto/pw
keys /etc/ntp/keys
disable monitor
fudge stratum 10

...and here's the sanitized config on the clients:

driftfile /var/lib/ntp/drift
restrict default nomodify notrap nopeer noquery
restrict ::1
restrict mask nomodify notrap
server ntp-server.home.local
includefile /etc/ntp/crypto/pw
keys /etc/ntp/keys
disable monitor

At this point, I'm not sure what the issue is, I know it isn't
connectivity, but I'm also not 100% confident in my configurations.
Thanks for taking a look, I would surely appreciate any help anyone
would be able to give.

More information about the questions mailing list