[NTLUG:Discuss] CentOS server not keeping time

[Rick Renshaw]

It looks like it could be a connectivity issue.  The reach column in your ntpq -p indicates the reachability of the remote servers.  It's an octal number that just keeps track of the last 8 connection attempts, and puts a 1 for success and 0 for failure into each bit of the octal number.  Right after the ntp service comes up, they start at 1, (00000001) and start moving up from there (i.e. 3 (00000011), then 7 (00000111)).  If the ntp service has been running a while, and you don't have any connectivity problems, reach will be 377 (11111111).  Your reach is still 0 on all the servers listed, which means all 8 of the last attempts to connect have failed.  Refid of .INIT. also supports that theory, as the refid of a connection is .INIT. until it makes the first connection and can get the refid from that server.

Here's what my ntpq -p looks like right now:

     remote           refid      st t when poll reach   delay   offset  jitter
==============================================================================
-50.7.0.147      147.231.2.6      2 u   27   64  377   55.587    3.736   2.912
+169.237.206.190 128.120.33.11    3 u   23   64   43   62.060   -1.774   1.615
*pacific.latt.ne 241.199.164.101  2 u   20   64  377   56.587   -3.571  13.812
+paladin.latt.ne 204.123.2.72     2 u  161  256  376   53.947   -3.076   4.299

You can see the first and third servers are great, with 377 reach.  The last one missed it's last attempt so it's 376 (1111110) and the second one isn't that reliable at all 43 (00100011) which means it missed 2, got 1, missed 3 more, and then made the last 2.

Rick



On Thursday, July 3, 2014 10:20 AM, Neil Aggarwal <neil at JAMMConsulting.com> wrote:
 


Hey everyone:

I have some additional information:

With ntpd running, if I do this command:
ntpdate -u 0.centos.pool.ntp.org
the time goes back to normal so it is obviously not a firewall or
connectivity
issue.  I did that a few days back and then the time is off again so I think
this is only a temporary fix.

If I do ntpq -p, I get this output:
     remote           refid      st t when poll reach   delay   offset
jitter
============================================================================
==
helium.constant .INIT.          16 u    - 1024    0    0.000    0.000
0.000
lithium.constan .INIT.          16 u    - 1024    0    0.000    0.000
0.000
bindcat.fhsu.ed .INIT.          16 u    - 1024    0    0.000    0.000
0.000
76.73.0.4       .INIT.          16 u    - 1024    0    0.000    0.000
0.000

It looks like the servers are still initializing, but my system has been
running
for quite some time  update gives me 39 min.  That should be enough time
for the servers to sync.

Any idea why the server is not syncing?

Thanks,
  Neil

--
Neil Aggarwal, (972) 834-1565
We lend money to investors to buy or refinance single family rent houses.
No origination fees, quick approval, no credit check.




_______________________________________________
http://www.ntlug.org/mailman/listinfo/discuss