[ntp:questions] How the "peer candidate" is selected?

pateretou Pateretou pateretou at hotmail.com
Fri Oct 22 16:37:00 UTC 2010



Hello,
I'm so embarrassing with this stupid question but I'm not very comfortable with the "peer select algorithm"

Let me try to explain:
I've got one GPS stratum 1 server located in New York (172.17.200.100) and one GPS stratum 1 server located in London (172.17.100.100).
in each datacenter approx 200 servers sync on BOTH servers

ntp.conf:
server 172.17.100.100 minpoll 4 maxpoll 4
server 172.17.200.100 minpoll 4 maxpoll 4

I would like servers in london sync to the London GPS and servers in NY sync with the NY GPS (seems to be a coherent choice, isn't it?)

The problem is all my servers are the NY server in "sys.peer". But the only "difference" between my timeserver with my servers in london is the delay is bigger with the NY server so I don't understand why my server use the NY server for sync and not the london server

$ ntpq -pn
     remote           refid      st t when poll reach   delay   offset  jitter
==============================================================================
+172.17.100.100     .PPS.            1 u    9   16  377    0.178    0.007   0.008
*172.17.200.100     .PPS.            1 u    8   16  377   67.203    0.009   0.010

Why my servers are sync with the most away time server?

# ntpq -c asso

ind assID status  conf reach auth condition  last_event cnt
===========================================================
  1 58141  9414   yes   yes  none  candidat   reachable  1
  2 58142  9614   yes   yes  none  sys.peer   reachable  1

the London timerserver:

# ntpq -c 'rv 58141'
assID=58141 status=9414 reach, conf, sel_candidat, 1 event, event_reach,
srcadr=172.17.100.100, srcport=123, dstadr=172.17.100.22, dstport=123,
leap=00, stratum=1, precision=-19, rootdelay=0.000,
rootdispersion=0.488, refid=PPS, reach=377, unreach=0, hmode=3, pmode=4,
hpoll=4, ppoll=4, flash=00 ok, keyid=0, ttl=0, offset=0.005,
delay=0.217, dispersion=0.488, jitter=0.029,
reftime=d06c3a1d.d2b37623  Fri, Oct 22 2010 17:28:13.823,
org=d06c3a2e.16e347ef  Fri, Oct 22 2010 17:28:30.089,
rec=d06c3a2e.16e982c5  Fri, Oct 22 2010 17:28:30.089,
xmt=d06c3a2e.16d6b6f5  Fri, Oct 22 2010 17:28:30.089,
filtdelay=     0.22    0.22    0.27    0.27    0.31    0.23    0.26    0.23,
filtoffset=    0.02    0.01   -0.04   -0.04   -0.01    0.02   -0.01    0.02,
filtdisp=      0.00    0.26    0.51    0.77    0.99    1.23    1.47    1.74

Ny Timeserver:
# ntpq -c 'rv 58142'
assID=58142 status=9614 reach, conf, sel_sys.peer, 1 event, event_reach,
srcadr=172.17.200.100, srcport=123, dstadr=172.16.100.22, dstport=123,
leap=00, stratum=1, precision=-20, rootdelay=0.000,
rootdispersion=0.275, refid=PPS, reach=377, unreach=0, hmode=3, pmode=4,
hpoll=4, ppoll=4, flash=00 ok, keyid=0, ttl=0, offset=0.013,
delay=67.217, dispersion=1.042, jitter=0.031,
reftime=d06c3a2e.c4b5e5f1  Fri, Oct 22 2010 17:28:30.768,
org=d06c3a32.1f8340a8  Fri, Oct 22 2010 17:28:34.123,
rec=d06c3a32.28203d08  Fri, Oct 22 2010 17:28:34.156,
xmt=d06c3a32.16e25ead  Fri, Oct 22 2010 17:28:34.089,
filtdelay=    67.29   67.24   67.30   67.37   67.32   67.24   67.22   67.37,
filtoffset=    0.00   -0.03   -0.01   -0.04   -0.02   -0.01    0.01    0.02,
filtdisp=      0.00    0.24    0.48    0.75    1.02    1.28    1.55    1.77


Do you have an idea how I can investigate on this issue?

Regards,
 		 	   		  


More information about the questions mailing list