NTP - poor reach after 20 hours

johnpoz

Yeah reach should be 377

You could do the math on a 200 to figure out what packets you have missed in the last 8

https://www.linuxjournal.com/article/6812

that they are all the same, would assume you had some sort of blip/bump/glitch where you were having connectivity issues for a small bit.

Are you saying it never goes over 200? Did you maybe just fill up your pipe or something.. Any sort of say saturation of your connection could cause reach issues..

farmerjohn

@johnpoz the reach does vary, but most of the time it is 1-3 packets out of 8.

ntpq -p
     remote           refid      st t when poll reach   delay   offset  jitter
==============================================================================
 0.us.pool.ntp.o .POOL.          16 p    -   64    0    0.000   +0.000   0.000
 1.us.pool.ntp.o .POOL.          16 p    -   64    0    0.000   +0.000   0.000
 2.us.pool.ntp.o .POOL.          16 p    -   64    0    0.000   +0.000   0.000
 3.us.pool.ntp.o .POOL.          16 p    -   64    0    0.000   +0.000   0.000
 tock.chi1.ntfo. 206.55.64.77     3 u   92  512    1   20.009  +10.665   6.826
*12.167.151.1    66.85.78.80      3 u  712  512    2   43.632   +1.343   1.403
+clock.fmt.he.ne .CDMA.           1 u  671  512    6   61.980   +0.363   2.513
 206-55-191-142. .PPS.            1 u  166  512    3   27.886   +0.876   1.838

johnpoz

Yeah that is not uncommon if your heavy use of your internet connection, or just peering problems..

I am in the middle of a video conference for work, and had been saturating my pipe before that with downloads.. So mine is prob not best example currently.

And I just updated to 21.02.2 last night via a clean install, etc..

You have an active peer, so you should be syncing time... I would just keep an eye on it.. Will do the same on mine, etc. But yeah when everything is fine.. you would want 377 for all of them..

bingo600

Offset & Jitter is terrible too.

Are the packets sent by Pigdin

/Bingo

johnpoz

20 hours is not really a long time when it comes to ntp ;)

Mine after the clean install has a pretty big offset its still working through... Seeing some steps in the log for ntp..

I would prob give it a few days.. Then take a look again.. That is what I am doing with mine..

Give it time to settle down..

Is this a new instance of pfsense? Has recently been rebooted, etc.. Or has it been up for days arleady?

farmerjohn

@johnpoz pfsense has been up for two weeks, but changed NTP to four pool servers from five individual servers from this NTP server list yesterday. I'll wait a few more days to see how it settles in.

johnpoz

You prob don't need more than 1 pool ;)

Yeah I am keeping an eye mind as well, since this is a clean install.. I am not sure if anything changed with the ntp client in this update, etc.

Will follow back in this in a few days.. I am seeing the same thing where reach is just not constant 377.. But also seeing the steps happening the log.. Which will refresh the reach count, etc.

example:

Apr 15 10:46:10 	ntpd 	42851 	0.0.0.0 0615 05 clock_sync
Apr 15 10:46:09 	ntpd 	42851 	0.0.0.0 061c 0c clock_step +0.475341 s

And offset is horrible currently..

Prob doesn't help that I just also updated and rebooted my local ntp server ;) But yeah ntp can take a while to setting down..

farmerjohn

@johnpoz said in NTP - poor reach after 20 hours:

You prob don't need more than 1 pool ;)

shoud I go ahead and delete the other 3 pools that I have and just use 0.us.pool.ntp.org?

johnpoz

Well you should prob use the pfsense pool ;)

The ntp pool allows for vendors to get their own pool names.. And users using pools using the vendors devices should use them... I haven't gotten around to changing mine - doing it now ;)

0.pfsense.pool.ntp.org

for example..
Just changed mine

here is info on the vendor pools
https://www.pool.ntp.org/en/vendors.html

bingo600

This is my linux server

And i haven't even sync'ed it up to my Tbolt or Samsung or ....
Just using inet peers.

/Bingo

ntpq -p
     remote           refid      st t when poll reach   delay   offset  jitter
==============================================================================
-n1.taur.dk      .PPS.            1 u  501 1024  377    7.190    0.022   0.841
+80.71.132.103 ( .GPS.            1 u  868 1024  377    1.821    0.268   0.135
*mmo2.ntp.se     .PPS.            1 u  971 1024  377    1.972    0.206   0.078
+78.156.103.10   217.198.219.102  2 u   77 1024  377    7.466    0.240   0.247
+mail.roostervan 212.99.225.86    2 u   16 1024  377    2.453    0.278   0.100
-time.cloudflare 10.20.10.78      3 u 1008 1024  377   10.140    0.042   0.113

johnpoz

^ yeah this a normal sort of status you should be seeing.

edit: That is some really low delays.. nice!

bingo600

@johnpoz

My pfSense

I use pfSense pool for pfSense.

But selected servers for my Linux.
n1.taur.dk is a friend of mine (also a time-nut) , his server is always "spot on"
My linux usually switches between n1.taur & the swedish one.

/Bingo

johnpoz

yeah those are some really nice low delays - have a nice internet connection ;)

farmerjohn

@johnpoz
switched over to pfsense.pool 90 min. ago. Here is status so far:

[2.5.0-RELEASE][admin@pfsense.home]/var/log: ntpq -p
     remote           refid      st t when poll reach   delay   offset  jitter
==============================================================================
 0.pfsense.pool. .POOL.          16 p    -   64    0    0.000   +0.000   0.000
 time.nullrouten 132.163.97.1     2 u  713  256  344   62.225   +0.955   0.504
+tick.nde.unlv.e 98.150.140.243   2 u  390  128  204   71.450   -7.756   2.534
+165.227.106.11  200.98.196.212   2 u  673  128  240   40.571   -0.168   3.267
*vps1.n1.ca      216.218.192.202  2 u  981  128  200   69.463   +1.543   2.424

Cool_Corona

bingo600

@cool_corona

There's quite a bit of jitter on your selected peer.
It' only won because it's a Stratum 1.

/Bingo

farmerjohn

@bingo600 subject of thread is 'poor' reach not 'porn' reach.

farmerjohn

@johnpoz
i updated from 2.5.0 to 2.5.1. Here is NTP status after 90 minutes after reboot. Hopefully the Scandinavian guys won't try and shame me again with their impressive NTP output.

[2.5.1-RELEASE][admin@pfsense.home]/var/log: ntpq -pcrv
     remote           refid      st t when poll reach   delay   offset  jitter
==============================================================================
 0.pfsense.pool. .POOL.          16 p    -   64    0    0.000   +0.000   0.000
+unifi.versadns. 71.66.197.233    2 u  632  128  120   42.354   -2.973   1.580
*clock.nyc.he.ne .CDMA.           1 u  453  128  144   37.191   +1.439   1.048

associd=0 status=0618 leap_none, sync_ntp, 1 event, no_sys_peer,
version="ntpd 4.2.8p15@1.3728-o Fri Feb  5 22:07:56 UTC 2021 (1)",
processor="amd64", system="FreeBSD/12.2-STABLE", leap=00, stratum=2,
precision=-21, rootdelay=37.191, rootdisp=476.256, refid=209.51.161.238,
reftime=e4233dba.ae57b7eb  Thu, Apr 15 2021 18:33:30.681,
clock=e42344a7.e24f1d1b  Thu, Apr 15 2021 19:03:03.884, peer=60391, tc=7,
mintc=3, offset=+1.048514, frequency=+38.158, sys_jitter=4.679301,
clk_jitter=1.975, clk_wander=0.060

Cool_Corona

A couple of the connections we have.

Cool_Corona

Try changing pools

Changed mine from dk, se, no to 0,1,2,3.europe.pool.ntp.org

It bettered the jitter quite considerably.