Re: [chrony-users] All network sources unusable

2023-06-05 Thread Torsten Wolf 
Apologies, Denny! I am unhappy myself that I do not respect the simplest 
mailing list etiquette such as fixed width font, sensitive line wrap and 
encoding since I switched to MacOS and Postbox. I guess it's time to get 
mutt up and running like in the old Linux days... Back on topic - I hope 
you can read Miroslav's reply and my response with the solution of the 
issue.


Many thanks for your support.

Denny Page schrieb am 04.06.23 um 16:53:
I’m sorry, unlike your prior email, I am unable to read the output in 
your last email.


Your prior email looked like this:
——
$ chronyc sourcestats
Name/IP Address    NP  NR  Span  Frequency  Freq Skew  Offset  
Std Dev

==
GPS 8   5   110   +444.701    259.006    +16ms  4453us
PPS 64  37  1011 -0.000  0.002 -1ns  1058ns
zeit.fu-berlin.de  0   0 0 +0.000 
2000.000 +0ns  4000ms
ptbtime1.ptb.de  0   0 0 +0.000   
2000.000 +0ns  4000ms
ptbtime2.ptb.de  0   0 0 +0.000 
2000.000 +0ns  4000ms
ptbtime3.ptb.de  0   0 0 +0.000   
2000.000 +0ns  4000ms
ns.tu-berlin.de  0   0 0 +0.000 
2000.000 +0ns  4000ms

——


Your new email looks like this:
——
MS Name/IP address Stratum Poll Reach LastRx Last sample
===
#? GPS聽聽 0聽聽 4聽聽 
377 8聽聽聽 -88ms[聽 -88ms] +/- 100ms
#* PPS聽聽 0聽聽 4聽聽 
113聽聽聽 14聽 -2398ns[-3176ns] +/- 1053ns
^? zeit.fu-berlin.de  
0聽聽 9聽聽 377 - +0ns[聽聽 +0ns] +/- 0ns
^? ptbtime1.ptb.de 聽聽 
0聽聽 9聽聽 377 - +0ns[聽聽 +0ns] +/- 0ns
^? ptbtime2.ptb.de 聽聽 
0聽聽 8聽聽 377 - +0ns[聽聽 +0ns] +/- 0ns
^? ptbtime3.ptb.de 聽聽 
0聽聽 8聽聽 377 - +0ns[聽聽 +0ns] +/- 0ns
^? ns.tu-berlin.de 聽聽 
0聽聽 8聽聽 377 - +0ns[聽聽 +0ns] +/- 0ns

Name/IP Address NP聽 NR聽 Span聽 Frequency聽 Freq Skew聽 Offset聽 Std Dev
==
GPS 64聽 50聽 1012 +5.202 36.190聽聽 -117ms聽聽聽 24ms
PPS 15聽聽 9聽聽 796 -0.000聽 0.006 -1ns聽 1630ns
zeit.fu-berlin.de  0聽聽 0 0 
+0.000聽聽 2000.000 +0ns聽 4000ms
ptbtime1.ptb.de  0聽聽 0 0 
+0.000聽聽 2000.000 +0ns聽 4000ms
ptbtime2.ptb.de  0聽聽 0 0 
+0.000聽聽 2000.000 +0ns聽 4000ms
ptbtime3.ptb.de  0聽聽 0 0 
+0.000聽聽 2000.000 +0ns聽 4000ms
ns.tu-berlin.de  0聽聽 0 0 
+0.000聽聽 2000.000 +0ns聽 4000ms

——


What you are expecting to see looks something like this:
——
MS Name/IP address         Stratum Poll Reach LastRx Last sample
===
^+ 192.168.230.201               1  -6   377     0    -10ns[  -10ns] 
+/- 13us
^* 192.168.230.202               1  -6   377     0  -11ns[  -12ns] +/- 
  13us
^? 192.168.230.203               1  -6   377     0    -11ns[  -11ns] 
+/-   13us
^? 192.168.230.200               1  -6   377     0    -52ns[  -52ns] 
+/- 12us
=? 192.168.230.11                2  -6   377     0 +5ns[   +5ns] +/-   
35us
^? 192.168.230.1                 2   3   377     7    +10us[  +10us] 
+/- 1121us

Name/IP Address      NP  NR  Span  Frequency  Freq Skew  Offset  Std Dev
==
192.168.230.201             7   5     1     -0.009      0.048    -13ns 
    9ns
192.168.230.202            33  17     8     -0.000      0.002    +10ns 
    7ns
192.168.230.203            50  27     1     +0.005      0.039     -9ns 
   19ns
192.168.230.200            64  36     1     -0.002      0.032    -77ns 
   21ns
192.168.230.11             37  17     1     -0.009      0.038    +14ns 
   11ns
192.168.230.1              29  15   227     +0.132      0.046    +13us 
 4369ns

——

Your actual values for will be different of course, but at a minimum 
there should be non zero values present for Reach, Last Sample, NP and 
Span. A value of 0 for NP, 0 for Span, 2000.000 for Freq Skew and 
4000ms for Std Dev are clear indications that the source is not being 
tracked.



Denny




On Jun 4, 2023, at 03:07, Torsten Wolf  wrote:

I added "noselect" to the NMEA (SHM0) source and restarted chrony. 
After an hour the output is


$ chronyc -m sources sourcestats
MS Name/IP address Stratum Poll Reach LastRx Last sample
===
#? GPS聽聽 0聽聽 4聽聽 
377 8聽聽聽 -88ms[聽 -88ms] +/- 100ms
#* PPS聽聽 0聽聽 4聽聽

Re: [chrony-users] All network sources unusable

2023-06-05 Thread Torsten Wolf 
That was the hint I needed! The main router the RasPi is connected to 
also runs chrony and acts as the timeserver for all local network 
clients. However, the script managing that chrony instance had an option 
to redirect all NTP traffic through the router. Interesting that I did 
not find that rule in the router config. This is the RasPi after 
toggling that option off:


MS Name/IP address Stratum Poll Reach LastRx Last sample
===
#- GPS   0   4   377 9    -45ms[  -45ms] +/- 
100ms
#* PPS   0   4   377 8   -461ns[ -537ns] +/- 
648ns
^? zeit.fu-berlin.de 1   8   377    66  -4025us[-4025us] +/- 
58ms
^? ptbtime1.ptb.de   1   8   377    71  -3100us[-3100us] +/- 
9719us
^? ptbtime2.ptb.de   1   8   377    72  -2490us[-2490us] +/- 
8901us
^? ptbtime3.ptb.de   1   8   377    73  -4077us[-4077us] +/- 
9630us
^? ns.tu-berlin.de   2   8   377    69  -3061us[-3062us] +/- 
16ms


And maybe this also solves the issue that the router (using the RasPi as 
sole source) was never able to correct an offset of -200ms. If not, I 
will come back soon.


Thanks again for your help!

Miroslav Lichvar schrieb am 05.06.23 um 10:02:

On Sun, Jun 04, 2023 at 12:07:44PM +0200, Torsten Wolf wrote:

Not sure why the Reach value for the PPS source is 113. A few minuts later
that value increased to 372, just to go down again afterwards.

That might be due to changing offset between GPS and PPS. Try
adjusting the offset option to make it closer to the middle as
explained in the FAQ.


Local address   : 192.168.1.105 (C0A80169)
Leap status : Normal
Version : 4
Mode    : Server
Stratum : 2
Poll interval   : 7 (128 seconds)
Precision   : -25 (0.00030 seconds)
Root delay  : 0.000778 seconds
Root dispersion : 0.269699 seconds
Reference ID    : C0A80169 ()
Reference time  : Sun Jun 04 10:00:57 2023
Offset  : -0.142506495 seconds
Peer delay  : 0.000236495 seconds
Peer dispersion : 0.00898 seconds
Response time   : 0.000178912 seconds
Jitter asymmetry: +0.00
NTP tests   : 111 111 1110

The test D is failing. That's the loop detection. The Reference ID is
the same as the local address, i.e. the client is polling a server
which seems to be synchronized to the client. As this happens with all
NTP servers and the peer delay is very short, I think it might be a
firewall in local network redirecting NTP requests to itself.

Re: [chrony-users] All network sources unusable

2023-06-05 Thread Denny Page 
I’m sorry, unlike your prior email, I am unable to read the output in your last 
email.

Your prior email looked like this:
——
$ chronyc sourcestats
Name/IP AddressNP  NR  Span  Frequency  Freq Skew  Offset  Std Dev
==
GPS 8   5   110   +444.701259.006+16ms  4453us
PPS 64  37  1011 -0.000  0.002 -1ns  1058ns
zeit.fu-berlin.de  0   0 0 +0.000   2000.000 
+0ns  4000ms
ptbtime1.ptb.de  0   0 0 +0.000   2000.000 
+0ns  4000ms
ptbtime2.ptb.de  0   0 0 +0.000   2000.000 
+0ns  4000ms
ptbtime3.ptb.de  0   0 0 +0.000   2000.000 
+0ns  4000ms
ns.tu-berlin.de  0   0 0 +0.000   2000.000 
+0ns  4000ms
——


Your new email looks like this:
——
MS Name/IP address Stratum Poll Reach LastRx Last sample
===
#? GPS聽聽 0聽聽 4聽聽 377 8聽聽聽 -88ms[聽 -88ms] +/- 100ms
#* PPS聽聽 0聽聽 4聽聽 113聽聽聽 14聽 -2398ns[-3176ns] +/- 1053ns
^? zeit.fu-berlin.de  0聽聽 9聽聽 377 
- +0ns[聽聽 +0ns] +/- 0ns
^? ptbtime1.ptb.de 聽聽 0聽聽 9聽聽 377 
- +0ns[聽聽 +0ns] +/- 0ns
^? ptbtime2.ptb.de 聽聽 0聽聽 8聽聽 377 
- +0ns[聽聽 +0ns] +/- 0ns
^? ptbtime3.ptb.de 聽聽 0聽聽 8聽聽 377 
- +0ns[聽聽 +0ns] +/- 0ns
^? ns.tu-berlin.de 聽聽 0聽聽 8聽聽 377 
- +0ns[聽聽 +0ns] +/- 0ns
Name/IP Address NP聽 NR聽 Span聽 Frequency聽 Freq Skew聽 Offset聽 Std Dev
==
GPS 64聽 50聽 1012 +5.202 36.190聽聽 -117ms聽聽聽 24ms
PPS 15聽聽 9聽聽 796 -0.000聽 0.006 -1ns聽 1630ns
zeit.fu-berlin.de  0聽聽 0 0 +0.000聽聽 
2000.000 +0ns聽 4000ms
ptbtime1.ptb.de  0聽聽 0 0 +0.000聽聽 2000.000 
+0ns聽 4000ms
ptbtime2.ptb.de  0聽聽 0 0 +0.000聽聽 2000.000 
+0ns聽 4000ms
ptbtime3.ptb.de  0聽聽 0 0 +0.000聽聽 2000.000 
+0ns聽 4000ms
ns.tu-berlin.de  0聽聽 0 0 +0.000聽聽 2000.000 
+0ns聽 4000ms
——


What you are expecting to see looks something like this:
——
MS Name/IP address Stratum Poll Reach LastRx Last sample   
===
^+ 192.168.230.201   1  -6   377 0-10ns[  -10ns] +/-   13us
^* 192.168.230.202   1  -6   377 0-11ns[  -12ns] +/-   13us
^? 192.168.230.203   1  -6   377 0-11ns[  -11ns] +/-   13us
^? 192.168.230.200   1  -6   377 0-52ns[  -52ns] +/-   12us
=? 192.168.230.112  -6   377 0 +5ns[   +5ns] +/-   35us
^? 192.168.230.1 2   3   377 7+10us[  +10us] +/- 1121us
Name/IP AddressNP  NR  Span  Frequency  Freq Skew  Offset  Std Dev
==
192.168.230.201 7   5 1 -0.009  0.048-13ns 9ns
192.168.230.20233  17 8 -0.000  0.002+10ns 7ns
192.168.230.20350  27 1 +0.005  0.039 -9ns19ns
192.168.230.20064  36 1 -0.002  0.032-77ns21ns
192.168.230.11 37  17 1 -0.009  0.038+14ns11ns
192.168.230.1  29  15   227 +0.132  0.046+13us  4369ns
——

Your actual values for will be different of course, but at a minimum there 
should be non zero values present for Reach, Last Sample, NP and Span. A value 
of 0 for NP, 0 for Span, 2000.000 for Freq Skew and 4000ms for Std Dev are 
clear indications that the source is not being tracked.


Denny



> On Jun 4, 2023, at 03:07, Torsten Wolf  wrote:
> 
> I added "noselect" to the NMEA (SHM0) source and restarted chrony. After an 
> hour the output is
> 
> $ chronyc -m sources sourcestats
> MS Name/IP address Stratum Poll Reach LastRx Last sample
> ===
> #? GPS聽聽 0聽聽 4聽聽 377 8聽聽聽 -88ms[聽 -88ms] +/- 100ms
> #* PPS聽聽 0聽聽 4聽聽 113聽聽聽 14聽 -2398ns[-3176ns] +/- 
> 1053ns
> ^? zeit.fu-berlin.de 0聽聽 9聽聽 377 - +0ns[聽聽 +0ns] +/- 0ns
> ^? ptbtime1.ptb.de聽聽 0聽聽 9聽聽 377 - +0ns[聽聽 +0ns] +/- 0ns
> ^? ptbtime2.ptb.de聽聽 0聽聽 8聽聽 377 - +0ns[聽聽 +0ns] +/- 0ns
> ^? ptbtime3.ptb.de聽聽 0聽聽 8聽聽 377 - +0ns[聽聽 +0ns] +/- 0ns
> ^? ns.tu-berlin.de聽聽 0聽聽 8聽聽 377 -

Re: [chrony-users] All network sources unusable

2023-06-05 Thread Miroslav Lichvar 
On Sun, Jun 04, 2023 at 12:07:44PM +0200, Torsten Wolf wrote:
> Not sure why the Reach value for the PPS source is 113. A few minuts later
> that value increased to 372, just to go down again afterwards.

That might be due to changing offset between GPS and PPS. Try
adjusting the offset option to make it closer to the middle as
explained in the FAQ.

> Local address   : 192.168.1.105 (C0A80169)
> Leap status : Normal
> Version : 4
> Mode    : Server
> Stratum : 2
> Poll interval   : 7 (128 seconds)
> Precision   : -25 (0.00030 seconds)
> Root delay  : 0.000778 seconds
> Root dispersion : 0.269699 seconds
> Reference ID    : C0A80169 ()
> Reference time  : Sun Jun 04 10:00:57 2023
> Offset  : -0.142506495 seconds
> Peer delay  : 0.000236495 seconds
> Peer dispersion : 0.00898 seconds
> Response time   : 0.000178912 seconds
> Jitter asymmetry: +0.00
> NTP tests   : 111 111 1110

The test D is failing. That's the loop detection. The Reference ID is
the same as the local address, i.e. the client is polling a server
which seems to be synchronized to the client. As this happens with all
NTP servers and the peer delay is very short, I think it might be a
firewall in local network redirecting NTP requests to itself.

-- 
Miroslav Lichvar


-- 
To unsubscribe email chrony-users-requ...@chrony.tuxfamily.org 
with "unsubscribe" in the subject.
For help email chrony-users-requ...@chrony.tuxfamily.org 
with "help" in the subject.
Trouble?  Email listmas...@chrony.tuxfamily.org.

Re: [chrony-users] All network sources unusable

2023-06-04 Thread Torsten Wolf 
I added "noselect" to the NMEA (SHM0) source and restarted chrony. After 
an hour the output is


$ chronyc -m sources sourcestats
MS Name/IP address Stratum Poll Reach LastRx Last sample
===
#? GPS   0   4   377 8    -88ms[  -88ms] +/- 
100ms
#* PPS   0   4   113    14  -2398ns[-3176ns] +/- 
1053ns

^? zeit.fu-berlin.de 0   9   377 - +0ns[   +0ns] +/- 0ns
^? ptbtime1.ptb.de   0   9   377 - +0ns[   +0ns] +/- 0ns
^? ptbtime2.ptb.de   0   8   377 - +0ns[   +0ns] +/- 0ns
^? ptbtime3.ptb.de   0   8   377 - +0ns[   +0ns] +/- 0ns
^? ns.tu-berlin.de   0   8   377 - +0ns[   +0ns] +/- 0ns
Name/IP Address NP  NR  Span  Frequency  Freq Skew  Offset  Std Dev
==
GPS 64  50  1012 +5.202 36.190   -117ms    24ms
PPS 15   9   796 -0.000  0.006 -1ns  1630ns
zeit.fu-berlin.de 0   0 0 +0.000   2000.000 +0ns  4000ms
ptbtime1.ptb.de 0   0 0 +0.000   2000.000 +0ns  4000ms
ptbtime2.ptb.de 0   0 0 +0.000   2000.000 +0ns  4000ms
ptbtime3.ptb.de 0   0 0 +0.000   2000.000 +0ns  4000ms
ns.tu-berlin.de 0   0 0 +0.000   2000.000 +0ns  4000ms

As far as I understand 377 indicates that the network sources can be 
accessed, and the fact that the measurements.log file is constantly 
updated with readings from those sources shows that data is being received.


Not sure why the Reach value for the PPS source is 113. A few minuts 
later that value increased to 372, just to go down again afterwards.


Here is a the output of "chronyc ntpdata" from one of the external sources

Remote address  : 192.53.103.104 (C0356768)
Remote port : 123
Local address   : 192.168.1.105 (C0A80169)
Leap status : Normal
Version : 4
Mode    : Server
Stratum : 2
Poll interval   : 7 (128 seconds)
Precision   : -25 (0.00030 seconds)
Root delay  : 0.000778 seconds
Root dispersion : 0.269699 seconds
Reference ID    : C0A80169 ()
Reference time  : Sun Jun 04 10:00:57 2023
Offset  : -0.142506495 seconds
Peer delay  : 0.000236495 seconds
Peer dispersion : 0.00898 seconds
Response time   : 0.000178912 seconds
Jitter asymmetry: +0.00
NTP tests   : 111 111 1110
Interleaved : No
Authenticated   : No
TX timestamping : Kernel
RX timestamping : Kernel
Total TX    : 16
Total RX    : 15
Total valid RX  : 15



Denny Page schrieb am 04.06.23 um 00:20:

Try this:

   chronyc -m sources sourcestats

This will print the sources and then the source stats. I suspect that the other 
sources you have are not actually being polled or are not reachable. Check the 
“Reach” column in sources.

Denny

Re: [chrony-users] All network sources unusable

2023-06-03 Thread Denny Page 
Try this:

  chronyc -m sources sourcestats

This will print the sources and then the source stats. I suspect that the other 
sources you have are not actually being polled or are not reachable. Check the 
“Reach” column in sources.

Denny


> On Jun 3, 2023, at 14:31, Torsten Wolf  wrote:
> 
> Bryan, many thanks for the explanation and the suggestion regarding SHM0. My 
> only worry is that the sourcestats output is not offering much in terms of 
> offset of the network sources:
> 
> $ chronyc sourcestats
> Name/IP AddressNP  NR  Span  Frequency  Freq Skew  Offset  Std Dev
> ==
> GPS 8   5   110   +444.701259.006+16ms  4453us
> PPS 64  37  1011 -0.000  0.002 -1ns  1058ns
> zeit.fu-berlin.de 0   0 0 +0.000   2000.000 +0ns  4000ms
> ptbtime1.ptb.de 0   0 0 +0.000   2000.000 +0ns  4000ms
> ptbtime2.ptb.de 0   0 0 +0.000   2000.000 +0ns  4000ms
> ptbtime3.ptb.de 0   0 0 +0.000   2000.000 +0ns  4000ms
> ns.tu-berlin.de 0   0 0 +0.000   2000.000 +0ns  4000ms
> 
> Cheers,
> Torsten
> 
> Bryan Christianson schrieb am 03.06.23 um 23:18:
>> Hi Torsten
>> 
>> From the chrony.conf(5) manpage:
>> noselect Never select this source. This is useful for monitoring or with 
>> sources which are not very accurate, but are locked with a PPS refclock.
>> 
>> Your 'noselect' hosts are not viable sources as they will never be selected. 
>> They will however show up in the 'chronyc sourcestats' list with their 
>> measured offsets.
>> 
>> If you look at the gpsd link again, you will see they are displaying 
>> 'chronyc sourcestats', not 'sources' as you have posted.
>> 
>> Also, I suggest you set the SHM 0 refclock to noselect - it will still be 
>> used by the pps refclock to give a timestamp to the pps readings but will 
>> not participate in setting the clock.
>> 
>> I hope that helps.
>> 
>> Regards, Bryan
>> 
>> 
> 


--
To unsubscribe email chrony-users-requ...@chrony.tuxfamily.org
with "unsubscribe" in the subject.
For help email chrony-users-requ...@chrony.tuxfamily.org
with "help" in the subject.
Trouble?  Email listmas...@chrony.tuxfamily.org.

Re: [chrony-users] All network sources unusable

2023-06-03 Thread Torsten Wolf 
Bryan, many thanks for the explanation and the suggestion regarding 
SHM0. My only worry is that the sourcestats output is not offering much 
in terms of offset of the network sources:


$ chronyc sourcestats
Name/IP Address    NP  NR  Span  Frequency  Freq Skew  Offset  
Std Dev

==
GPS 8   5   110   +444.701    259.006    +16ms  4453us
PPS 64  37  1011 -0.000  0.002 -1ns  1058ns
zeit.fu-berlin.de 0   0 0 +0.000   2000.000 +0ns  4000ms
ptbtime1.ptb.de 0   0 0 +0.000   2000.000 +0ns  4000ms
ptbtime2.ptb.de 0   0 0 +0.000   2000.000 +0ns  4000ms
ptbtime3.ptb.de 0   0 0 +0.000   2000.000 +0ns  4000ms
ns.tu-berlin.de 0   0 0 +0.000   2000.000 +0ns  4000ms

Cheers,
Torsten

Bryan Christianson schrieb am 03.06.23 um 23:18:

Hi Torsten

 From the chrony.conf(5) manpage:
noselect Never select this source. This is useful for monitoring or with 
sources which are not very accurate, but are locked with a PPS refclock.

Your 'noselect' hosts are not viable sources as they will never be selected. 
They will however show up in the 'chronyc sourcestats' list with their measured 
offsets.

If you look at the gpsd link again, you will see they are displaying 'chronyc 
sourcestats', not 'sources' as you have posted.

Also, I suggest you set the SHM 0 refclock to noselect - it will still be used 
by the pps refclock to give a timestamp to the pps readings but will not 
participate in setting the clock.

I hope that helps.

Regards, Bryan






smime.p7s
Description: S/MIME Cryptographic Signature

Re: [chrony-users] All network sources unusable

2023-06-03 Thread Bryan Christianson 
Hi Torsten

From the chrony.conf(5) manpage: 
noselect Never select this source. This is useful for monitoring or with 
sources which are not very accurate, but are locked with a PPS refclock.

Your 'noselect' hosts are not viable sources as they will never be selected. 
They will however show up in the 'chronyc sourcestats' list with their measured 
offsets.

If you look at the gpsd link again, you will see they are displaying 'chronyc 
sourcestats', not 'sources' as you have posted.

Also, I suggest you set the SHM 0 refclock to noselect - it will still be used 
by the pps refclock to give a timestamp to the pps readings but will not 
participate in setting the clock.

I hope that helps.

Regards, Bryan


-- 
Bryan Christianson
br...@whatroute.net




--
To unsubscribe email chrony-users-requ...@chrony.tuxfamily.org
with "unsubscribe" in the subject.
For help email chrony-users-requ...@chrony.tuxfamily.org
with "help" in the subject.
Trouble?  Email listmas...@chrony.tuxfamily.org.