[ntp:questions] Strange behaviour

Alain alainm at pobox.com
Tue Mar 15 17:05:43 UTC 2005


John Pettitt escreveu:
 >>
 > Look for (reach == 0  for *all* non-local clocks)  and (LOCAL clock
 > reach == 377)
 >
 > this will detect that case when the system can't see the world but has
 > been up long enough to have tried  (the LOCAL == 377 test)

I have started some tests in that efect (to test ntpd sanity) and got 
very strange results. I restarted the internet connection in order to 
change my IP (something that can happen unexpectedly) ando now "ntpq -p" 
is showing duplicate servers!!!

This was little after changing my IP:
[root at servnet1 root]# ntpq -p
      remote           refid      st t when poll reach   delay   offset 
  jitter
==============================================================================
+ntp.pucpr.br    tick.ucla.edu    2 u  289  512  377  242.036    0.750 
392.381
-paulo.hiway.com Tick.UH.EDU      2 u  285  512  377  230.962  -38.051 
37.589
  LOCAL(0)        LOCAL(0)        10 l   40   64  377    0.000    0.000 
   0.004
-200.137.65.100  ntp1.pop-df.rnp  2 u  469  512  377  394.278  113.750 
  7.127
  titan.cais.rnp. 217.69.95.255    2 u  475  512  377  1497.11  665.964 
615.687
-cherokee.pop-pr ntp1.pop-df.rnp  2 u  279  512  377  176.529  -38.452 
22.191
xmailhost.pads.u santuario.pads.  2 u  591  512  366  260.148  -245.71 
686.320
*cs.columbia.edu clepsydra.dec.c  2 u  473  512  377  410.720   -2.734 
10.116
+hydrogen.cert.u tick.ucla.edu    2 u  466  512  377  437.483   -5.566 
28.325
-200.137.65.100  ntp1.pop-df.rnp  2 -   21   64    1  294.214  110.186 
18.106
  titan.cais.rnp. 152.53.71.255    2 -   38   64    1  238.410   84.797 
  51.639
-cherokee.pop-pr ntp1.pop-df.rnp  2 -   34   64    1  193.572   10.393 
70.913
xns.pads.ufrj.br ns2.pads.ufrj.b  2 -   26   64    1  251.871  -215.37 
22.417
-cs.columbia.edu clepsydra.dec.c  2 -   35   64    1  348.125   21.730 
21.164
-hydrogen.cert.u tick.ucla.edu    2 -   30   64    1  389.365   24.507 
24.961

Is this expected? What is happening?

thanks,
Alain
PS What is the column marked "t"? Where can I find a complete 
explanation of this listing?



More information about the questions mailing list