1

It is a just installed ubuntu 16.04.2 server, and I installed ntp package there, which currently runs with the default config with only single line added:

enable mode7

since otherwise ntpdc and collectd cannot fetch data from it.

And what I cannot interpret is this output:

# ntpdc -c peers localhost
     remote           local      st poll reach  delay   offset    disp
=======================================================================
=ntp2.ntp.net.nz 10.50.200.3      1  256  377 0.01117  0.000599 0.10858
*timeball1.its.w 10.50.200.3      1  256  377 0.01054 -0.000771 0.10974
=timeball3.its.w 10.50.200.3      1  256  377 0.01007 -0.001039 0.11723
=ns1.tdc.akl.tel 10.50.200.3      2  512  377 0.00882  0.000451 0.12932
=ntp1.ntp.net.nz 10.50.200.3      1  256  377 0.01041  0.000254 0.13625
=0.0.0.0         0.0.0.0         16   64    0 0.00000  0.000000 4.00000
=0.0.0.0         0.0.0.0         16   64    0 0.00000  0.000000 4.00000
=0.0.0.0         0.0.0.0         16   64    0 0.00000  0.000000 4.00000
=0.0.0.0         0.0.0.0         16   64    0 0.00000  0.000000 4.00000
=0.0.0.0         0.0.0.0         16   64    0 0.00000  0.000000 4.00000

What are those 5 lines with 0.0.0.0 remote?

Additionally, here is the output from ntpq:

# ntpq -p
     remote           refid      st t when poll reach   delay   offset  jitter
==============================================================================
 0.ubuntu.pool.n .POOL.          16 p    -   64    0    0.000    0.000   0.000
 1.ubuntu.pool.n .POOL.          16 p    -   64    0    0.000    0.000   0.000
 2.ubuntu.pool.n .POOL.          16 p    -   64    0    0.000    0.000   0.000
 3.ubuntu.pool.n .POOL.          16 p    -   64    0    0.000    0.000   0.000
 ntp.ubuntu.com  .POOL.          16 p    -   64    0    0.000    0.000   0.000
-ntp1.ntp.net.nz .GPS.            1 u   43  256  377   10.418    0.254   0.443
-ns1.tdc.akl.tel 74.189.58.78     2 u  351  512  377    8.831    0.451   0.270
+timeball3.its.w .GPS.            1 u  148  256  377   10.080   -1.039   1.199
*timeball1.its.w .GPS.            1 u   73  256  377   10.551   -0.771   0.431
+ntp2.ntp.net.nz .GPS.            1 u   79  256  377   11.183    0.599   0.270

Why this question at all:

collectd causes these annoying syslog messages and I think it's relevant:

Jul 12 01:59:45 server collectd[2773]: uc_update: Value too old: name = server.domain.tld/ntpd/time_dispersion-0.0.0.0; value time = 1499824785.998; last cache update = 1499824785.998;
Jul 12 01:59:45 server collectd[2773]: uc_update: Value too old: name = server.domain.tld/ntpd/time_offset-0.0.0.0; value time = 1499824785.998; last cache update = 1499824785.998;
Jul 12 01:59:45 server collectd[2773]: uc_update: Value too old: name = server.domain.tld/ntpd/delay-0.0.0.0; value time = 1499824785.998; last cache update = 1499824785.998;
0
1

Those are placeholders for the pool associations. See this bug:

Bug 2014: strange interaction between pool directive and maxclock

Notice the p in the type column? That indicates its a placeholder entry for pool directives.

As for "why this question" the problem lies in whatever you are using to parse the output before sending it to collectd. It should ignore any line that contains a type of p

3
  • tos maxclock 8 did not change anything. There is still 5 lines of code, exactly how many pool servers there are defined. If the number of pool servers changes - the number of those 0.0.0.0 lines changes accordingly. Sorry, not an answer. – zerkms Jul 12 '17 at 19:58
  • @zerkms i updated the answer. those entries are an artifact of the way ntp does pool associations. – dfc Jul 12 '17 at 20:23
  • "It should ignore any line that contains a type of p" --- yep, those were exactly my thoughts as well, hence I submitted an issue to the collectd repository: github.com/collectd/collectd/issues/2358 – zerkms Jul 12 '17 at 20:31

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.