[OmniOS-discuss] OmniOS ntpd multicast client oddity

Bob Friesenhahn bfriesen at simple.dallas.tx.us
Wed Jan 25 14:35:17 UTC 2017


Under Solaris 10 and OpenIndiana oi_151a9 I am successfully running 
ntp clients in multicast client mode using trusted keys for security. 
I tried to do the same on a fresh OmniOS r151020 install (on a ixgbe 
interface running at 1Gbit) but find that ntpd only works properly for 
a short while after being started or restarted (based on 'ntpq -p' 
output) but then something goes wrong.  After a while there is no more 
output from 'ntpq -p' and I see this with debug enabled:

ntpq -d -p
1 packets reassembled into response
[12794] [12793] [12792] [12791]
4 associations total
::1 reversed to localhost
      remote           refid      st t when poll reach   delay   offset  jitter
==============================================================================
eliding [12791]
eliding [12792]
eliding [12793]
eliding [12794]

I see this on the OpenIndiana host which is attached to the same 
switch:

% ntpq -n -d -p
1 packets reassembled into response
[12482] [12481] [12480] [12479] [12478]
5 associations total
      remote           refid      st t when poll reach   delay   offset  jitter
==============================================================================
eliding [12478]
2 packets reassembled into response
*65.66.246.73    204.9.54.119     2 b   43   64  376    0.000    0.422   0.087
2 packets reassembled into response
+65.66.246.74    132.239.1.6      2 b   48   64  376    0.000   -1.064   0.176
eliding [12481]
2 packets reassembled into response
+65.66.246.91    204.9.54.119     2 b   43   64  376    0.000    0.426   0.078

and this is what I see on the OmniOS host shortly after ntpd has been 
started/re-started:

  ntpq -n -d -p
1 packets reassembled into response
[62163] [62162] [62161] [62160]
4 associations total
      remote           refid      st t when poll reach   delay   offset  jitter
==============================================================================
2 packets reassembled into response
+65.66.246.74    132.239.1.6      2 b   12   64    1    0.862   -1.652   0.167
eliding [62161]
2 packets reassembled into response
*65.66.246.91    204.9.54.119     2 b    8   64    1    0.272   -0.554   0.035
2 packets reassembled into response
+65.66.246.73    204.9.54.119     2 b   11   64    1    0.272   -0.535   0.023

The key statement in the ntp.conf file is this:

# Passively wait for a server to provide NTP packets on the ntp multicast net.
multicastclient 224.0.1.1

I am wondering why all of the peer entries are being 'elided' after 
some time (maybe minutes).  Is multicast reception only working for a 
few packets and then ceases or is something else going on?

Thanks,

Bob
-- 
Bob Friesenhahn
bfriesen at simple.dallas.tx.us, http://www.simplesystems.org/users/bfriesen/
GraphicsMagick Maintainer,    http://www.GraphicsMagick.org/


More information about the OmniOS-discuss mailing list