Discussion:
[chrony-users] PPS correction not being applied
Brian Gieryk
2017-02-23 16:00:55 UTC
Permalink
Good morning.

Running chrony 3.1, raspberry Pi 3 model B with the latest Raspbian Jessie.

Have GPS working to set the clock with no issue.

Have good PPS on /dev/pps0

From searching the archives I have :
refclock PPS /dev/pps0 poll 0 refid PPS lock NMEA

Appended to the end of /etc/chrony/chrony.conf

Running chronyd produces no errors

PPS shows in chronyc sources

However the #? next to PPS never changes (48 hours so far) and there are no corrections.

Any thoughts?

Thank you all!

Brian

Sent from my iPhone
--
To unsubscribe email chrony-users-***@chrony.tuxfamily.org
with "unsubscribe" in the subject.
For help email chrony-users-***@chrony.tuxfamily.org
with "help" in the subject.
Trouble? Email ***@chrony.tuxfamily.org.
Bill Unruh
2017-02-23 16:09:25 UTC
Permalink
You tell it to lock to NMEA. Do you actually have an NMEA refclock running?
Ie you might want to show us what your complete chrony.conf is.

Also in your sources, what sources are listed?
Ie, what is the complete output of chronyc sources command.

In the logging of the refclock, are there any entries? Do they give you a
clue?



William G. Unruh __| Canadian Institute for|____ Tel: +1(604)822-3273
Physics&Astronomy _|___ Advanced Research _|____ Fax: +1(604)822-5324
UBC, Vancouver,BC _|_ Program in Cosmology |____ ***@physics.ubc.ca
Canada V6T 1Z1 ____|____ and Gravity ______|_ www.theory.physics.ubc.ca/
Post by Brian Gieryk
Good morning.
Running chrony 3.1, raspberry Pi 3 model B with the latest Raspbian Jessie.
Have GPS working to set the clock with no issue.
Have good PPS on /dev/pps0
refclock PPS /dev/pps0 poll 0 refid PPS lock NMEA
Appended to the end of /etc/chrony/chrony.conf
Running chronyd produces no errors
PPS shows in chronyc sources
However the #? next to PPS never changes (48 hours so far) and there are no corrections.
Any thoughts?
Thank you all!
Brian
Sent from my iPhone
--
with "unsubscribe" in the subject.
with "help" in the subject.
--
To unsubscribe email chrony-users-***@chrony.tuxfamily.org
with "unsubscribe" in the subject.
For help email chrony-users-***@chrony.tuxfamily.org
with "help" in the subject.
Trouble? Email ***@chrony.tuxfamily.org.
Brian Gieryk
2017-02-24 00:10:08 UTC
Permalink
Relevant chrony.conf snipet:

# offline sources

refclock SHM 0 refid GPS precision 1e-1
refclock SHM 1 refid NMEA precision 1e-9
refclock SOCK /tmp/chrony.ttyS0.sock refid NMEA
refclock PPS /dev/pps0 poll 0 refid PPS lock NMEA

Sources output:

***@raspberrypi:~ $ sudo chronyc sources
210 Number of sources = 5
MS Name/IP address Stratum Poll Reach LastRx Last
sample
===============================================================================
#* GPS 0 4 377 15-+1007us [-1476us]+/-100ms
#? NMEA 0 4 0 -+0ns [ +0ns]+/- 0ns
#? NMEA 0 4 0 -+0ns [ +0ns]+/- 0ns
#? PPS 0 0 0 -+0ns [ +0ns]+/- 0ns
^? 2610:20:6f15:15::27 0 8 0 -+0ns [ +0ns]+/- 0ns
***@raspberrypi:~ $

I don't see anything unusual(?) in the log...

I am not an old hand at this......

Thank you!

Brian
Post by Bill Unruh
You tell it to lock to NMEA. Do you actually have an NMEA refclock running?
Ie you might want to show us what your complete chrony.conf is.
Also in your sources, what sources are listed? Ie, what is the complete output of chronyc sources command.
In the logging of the refclock, are there any entries? Do they give you a
clue?
William G. Unruh __| Canadian Institute for|____ Tel: +1(604)822-3273
Physics&Astronomy _|___ Advanced Research _|____ Fax: +1(604)822-5324
Canada V6T 1Z1 ____|____ and Gravity ______|_ www.theory.physics.ubc.ca/
Post by Brian Gieryk
Good morning.
Running chrony 3.1, raspberry Pi 3 model B with the latest Raspbian Jessie.
Have GPS working to set the clock with no issue.
Have good PPS on /dev/pps0
refclock PPS /dev/pps0 poll 0 refid PPS lock NMEA
Appended to the end of /etc/chrony/chrony.conf
Running chronyd produces no errors
PPS shows in chronyc sources
However the #? next to PPS never changes (48 hours so far) and there are no corrections.
Any thoughts?
Thank you all!
Brian
Sent from my iPhone
--
with "unsubscribe" in the subject.
with "help" in the subject.
--
--
To unsubscribe email chrony-users-***@chrony.tuxfamily.org
with "unsubscribe" in the subject.
For help email chrony-users-***@chrony.tuxfamily.org
with "help" in the subject.
Trouble? Email ***@chrony.tuxfamily.org.
Bill Unruh
2017-02-24 00:46:53 UTC
Permalink
You do not know what is relevant do you?
Post by Brian Gieryk
# offline sources
refclock SHM 0 refid GPS precision 1e-1
What is putting this info into shm?
Post by Brian Gieryk
refclock SHM 1 refid NMEA precision 1e-9
refclock SOCK /tmp/chrony.ttyS0.sock refid NMEA
Two different sources with the same name. That could cause confusion.
Where do these come from?
Post by Brian Gieryk
refclock PPS /dev/pps0 poll 0 refid PPS lock NMEA
210 Number of sources = 5
MS Name/IP address Stratum Poll Reach LastRx Last
sample
===============================================================================
#* GPS 0 4 377 15-+1007us [-1476us]+/-100ms
GPS reports. So why not lock to it?
Post by Brian Gieryk
#? NMEA 0 4 0 -+0ns [ +0ns]+/- 0ns
#? NMEA 0 4 0 -+0ns [ +0ns]+/- 0ns
Neither of your nmea sources report so PPS has nothing to lock to.
Post by Brian Gieryk
#? PPS 0 0 0 -+0ns [ +0ns]+/- 0ns
^? 2610:20:6f15:15::27 0 8 0 -+0ns [ +0ns]+/- 0ns
What is this source?
Post by Brian Gieryk
I don't see anything unusual(?) in the log...
Do you see anything? Why not give us 10 or 20 lines from the refclock log?
Post by Brian Gieryk
I am not an old hand at this......
Thank you!
Brian
Post by Bill Unruh
You tell it to lock to NMEA. Do you actually have an NMEA refclock running?
Ie you might want to show us what your complete chrony.conf is.
Also in your sources, what sources are listed? Ie, what is the complete output of chronyc sources command.
In the logging of the refclock, are there any entries? Do they give you a
clue?
William G. Unruh __| Canadian Institute for|____ Tel: +1(604)822-3273
Physics&Astronomy _|___ Advanced Research _|____ Fax: +1(604)822-5324
Canada V6T 1Z1 ____|____ and Gravity ______|_ www.theory.physics.ubc.ca/
Post by Brian Gieryk
Good morning.
Running chrony 3.1, raspberry Pi 3 model B with the latest Raspbian Jessie.
Have GPS working to set the clock with no issue.
Have good PPS on /dev/pps0
refclock PPS /dev/pps0 poll 0 refid PPS lock NMEA
Appended to the end of /etc/chrony/chrony.conf
Running chronyd produces no errors
PPS shows in chronyc sources
However the #? next to PPS never changes (48 hours so far) and there are no corrections.
Any thoughts?
Thank you all!
Brian
Sent from my iPhone
--
with "unsubscribe" in the subject.
with "help" in the subject.
--
--
with "unsubscribe" in the subject.
with "help" in the subject.
--
To unsubscribe email chrony-users-***@chrony.tuxfamily.org
with "unsubscribe" in the subject.
For help email chrony-users-***@chrony.tuxfamily.org
with "help" in the subject.
Trouble? Email ***@chrony.tuxfamily.org.
Brian Gieryk
2017-02-24 01:09:54 UTC
Permalink
I apologize for my seeming ignorance, I have been trying everything I could find in the archives here.

I have been using Linux for a grand total of one week, so some patience will be much appreciated.

My Pi does not seem to send email directly to anyone other than myself, so it is a 2 step process to send anything from it.

As for locking to the GPS, the time/date are being set correctly from it now, I would just like some redundancy, and a tad more accuracy, if it can be had.

I will get more info shortly.

Thank you for your understanding.

Brian

Sent from my iPhone
Post by Bill Unruh
You do not know what is relevant do you?
Post by Brian Gieryk
# offline sources
refclock SHM 0 refid GPS precision 1e-1
What is putting this info into shm?
Post by Brian Gieryk
refclock SHM 1 refid NMEA precision 1e-9
refclock SOCK /tmp/chrony.ttyS0.sock refid NMEA
Two different sources with the same name. That could cause confusion. Where do these come from?
Post by Brian Gieryk
refclock PPS /dev/pps0 poll 0 refid PPS lock NMEA
210 Number of sources = 5
MS Name/IP address Stratum Poll Reach LastRx Last
sample
===============================================================================
#* GPS 0 4 377 15-+1007us [-1476us]+/-100ms
GPS reports. So why not lock to it?
Post by Brian Gieryk
#? NMEA 0 4 0 -+0ns [ +0ns]+/- 0ns
#? NMEA 0 4 0 -+0ns [ +0ns]+/- 0ns
Neither of your nmea sources report so PPS has nothing to lock to.
Post by Brian Gieryk
#? PPS 0 0 0 -+0ns [ +0ns]+/- 0ns
^? 2610:20:6f15:15::27 0 8 0 -+0ns [ +0ns]+/- 0ns
What is this source?
Post by Brian Gieryk
I don't see anything unusual(?) in the log...
Do you see anything? Why not give us 10 or 20 lines from the refclock log?
Post by Brian Gieryk
I am not an old hand at this......
Thank you!
Brian
Post by Bill Unruh
You tell it to lock to NMEA. Do you actually have an NMEA refclock running?
Ie you might want to show us what your complete chrony.conf is.
Also in your sources, what sources are listed? Ie, what is the complete output of chronyc sources command.
In the logging of the refclock, are there any entries? Do they give you a
clue?
William G. Unruh __| Canadian Institute for|____ Tel: +1(604)822-3273
Physics&Astronomy _|___ Advanced Research _|____ Fax: +1(604)822-5324
Canada V6T 1Z1 ____|____ and Gravity ______|_ www.theory.physics.ubc.ca/
Post by Brian Gieryk
Good morning.
Running chrony 3.1, raspberry Pi 3 model B with the latest Raspbian Jessie.
Have GPS working to set the clock with no issue.
Have good PPS on /dev/pps0
refclock PPS /dev/pps0 poll 0 refid PPS lock NMEA
Appended to the end of /etc/chrony/chrony.conf
Running chronyd produces no errors
PPS shows in chronyc sources
However the #? next to PPS never changes (48 hours so far) and there are no corrections.
Any thoughts?
Thank you all!
Brian
Sent from my iPhone
--
with "unsubscribe" in the subject.
with "help" in the subject.
--
--
with "unsubscribe" in the subject.
with "help" in the subject.
--
--
To unsubscribe email chrony-users-***@chrony.tuxfamily.org
with "unsubscribe" in the subject.
For help email chrony-users-***@chrony.tuxfamily.org
with "help" in the subject.
Trouble? Email ***@chrony.tuxfamily.org.
Bill Unruh
2017-02-24 07:43:45 UTC
Permalink
Post by Brian Gieryk
I apologize for my seeming ignorance, I have been trying everything I could find in the archives here.
No problem. JUst remember that we have no information about your system except
what you tell us.
Post by Brian Gieryk
I have been using Linux for a grand total of one week, so some patience will be much appreciated.
My Pi does not seem to send email directly to anyone other than myself, so it is a 2 step process to send anything from it.
As for locking to the GPS, the time/date are being set correctly from it now, I would just like some redundancy, and a tad more accuracy, if it can be had.
But I am a bit confused. Do you have two or 3 GPS receivers attached to
your Pi? Or are all of GPS, the two NMEA and PPS coming (as I suspect)
from one GPS receiver. If my suspicion is true, then that is not redundancy.
The various attempts t read 3 different times from the same receiver could
actually make things worse, not better (the readings fighting with each other).

Secondly try to get one thing going at a time. GPS, whatever it is (what
reports the gps to the SHM?-- are you using gpsd?) seems to work now. THe NMEA
are not. Use what works to gine the PPS its seconds-resolution, so PPS can
deliver its soperior accuracy, and so you can see if you can get it working.
Worry about redundancy later.
Post by Brian Gieryk
I will get more info shortly.
Thank you for your understanding.
Brian
Sent from my iPhone
Post by Bill Unruh
You do not know what is relevant do you?
Post by Brian Gieryk
# offline sources
refclock SHM 0 refid GPS precision 1e-1
What is putting this info into shm?
Post by Brian Gieryk
refclock SHM 1 refid NMEA precision 1e-9
refclock SOCK /tmp/chrony.ttyS0.sock refid NMEA
Two different sources with the same name. That could cause confusion. Where do these come from?
Post by Brian Gieryk
refclock PPS /dev/pps0 poll 0 refid PPS lock NMEA
210 Number of sources = 5
MS Name/IP address Stratum Poll Reach LastRx Last
sample
===============================================================================
#* GPS 0 4 377 15-+1007us [-1476us]+/-100ms
GPS reports. So why not lock to it?
Post by Brian Gieryk
#? NMEA 0 4 0 -+0ns [ +0ns]+/- 0ns
#? NMEA 0 4 0 -+0ns [ +0ns]+/- 0ns
Neither of your nmea sources report so PPS has nothing to lock to.
Post by Brian Gieryk
#? PPS 0 0 0 -+0ns [ +0ns]+/- 0ns
^? 2610:20:6f15:15::27 0 8 0 -+0ns [ +0ns]+/- 0ns
What is this source?
Post by Brian Gieryk
I don't see anything unusual(?) in the log...
Do you see anything? Why not give us 10 or 20 lines from the refclock log?
Post by Brian Gieryk
I am not an old hand at this......
Thank you!
Brian
Post by Bill Unruh
You tell it to lock to NMEA. Do you actually have an NMEA refclock running?
Ie you might want to show us what your complete chrony.conf is.
Also in your sources, what sources are listed? Ie, what is the complete output of chronyc sources command.
In the logging of the refclock, are there any entries? Do they give you a
clue?
William G. Unruh __| Canadian Institute for|____ Tel: +1(604)822-3273
Physics&Astronomy _|___ Advanced Research _|____ Fax: +1(604)822-5324
Canada V6T 1Z1 ____|____ and Gravity ______|_ www.theory.physics.ubc.ca/
Post by Brian Gieryk
Good morning.
Running chrony 3.1, raspberry Pi 3 model B with the latest Raspbian Jessie.
Have GPS working to set the clock with no issue.
Have good PPS on /dev/pps0
refclock PPS /dev/pps0 poll 0 refid PPS lock NMEA
Appended to the end of /etc/chrony/chrony.conf
Running chronyd produces no errors
PPS shows in chronyc sources
However the #? next to PPS never changes (48 hours so far) and there are no corrections.
Any thoughts?
Thank you all!
Brian
Sent from my iPhone
--
with "unsubscribe" in the subject.
with "help" in the subject.
--
--
with "unsubscribe" in the subject.
with "help" in the subject.
--
--
with "unsubscribe" in the subject.
with "help" in the subject.
--
To unsubscribe email chrony-users-***@chrony.tuxfamily.org
with "unsubscribe" in the subject.
For help email chrony-users-***@chrony.tuxfamily.org
with "help" in the subject.
Trouble? Email ***@chrony.tuxfamily.org.
Brian Gieryk
2017-02-24 15:01:16 UTC
Permalink
Ok, there is one GPS (with PPS out on a second output, going to GPIO 18 on the rpi)

I did a little change (commented out the two NMEA sources.

I am running gpsd for the GPS source.

It is a Dragino LoRaWAN/GPS hat.

As I mentioned, the GPS is functioning (what appears to me) correctly.

I will add refclocks logging right now.

I will most likely add a second GPS, or more probably, tie it into my Rubidium standard.

As for the PPS signal, ppstest /dev/pps0 give the expected output, finds the PPS signal, and is happy with it.

Chronyc sources just does not seem to recognize it.

Thank you!

Brian

Sent from my iPhone
Post by Bill Unruh
Post by Brian Gieryk
I apologize for my seeming ignorance, I have been trying everything I could find in the archives here.
No problem. JUst remember that we have no information about your system except
what you tell us.
Post by Brian Gieryk
I have been using Linux for a grand total of one week, so some patience will be much appreciated.
My Pi does not seem to send email directly to anyone other than myself, so it is a 2 step process to send anything from it.
As for locking to the GPS, the time/date are being set correctly from it now, I would just like some redundancy, and a tad more accuracy, if it can be had.
But I am a bit confused. Do you have two or 3 GPS receivers attached to
your Pi? Or are all of GPS, the two NMEA and PPS coming (as I suspect) from one GPS receiver. If my suspicion is true, then that is not redundancy.
The various attempts t read 3 different times from the same receiver could
actually make things worse, not better (the readings fighting with each other).
Secondly try to get one thing going at a time. GPS, whatever it is (what
reports the gps to the SHM?-- are you using gpsd?) seems to work now. THe NMEA
are not. Use what works to gine the PPS its seconds-resolution, so PPS can
deliver its soperior accuracy, and so you can see if you can get it working.
Worry about redundancy later.
Post by Brian Gieryk
I will get more info shortly.
Thank you for your understanding.
Brian
Sent from my iPhone
Post by Bill Unruh
You do not know what is relevant do you?
Post by Brian Gieryk
# offline sources
refclock SHM 0 refid GPS precision 1e-1
What is putting this info into shm?
Post by Brian Gieryk
refclock SHM 1 refid NMEA precision 1e-9
refclock SOCK /tmp/chrony.ttyS0.sock refid NMEA
Two different sources with the same name. That could cause confusion. Where do these come from?
Post by Brian Gieryk
refclock PPS /dev/pps0 poll 0 refid PPS lock NMEA
210 Number of sources = 5
MS Name/IP address Stratum Poll Reach LastRx Last
sample
===============================================================================
#* GPS 0 4 377 15-+1007us [-1476us]+/-100ms
GPS reports. So why not lock to it?
Post by Brian Gieryk
#? NMEA 0 4 0 -+0ns [ +0ns]+/- 0ns
#? NMEA 0 4 0 -+0ns [ +0ns]+/- 0ns
Neither of your nmea sources report so PPS has nothing to lock to.
Post by Brian Gieryk
#? PPS 0 0 0 -+0ns [ +0ns]+/- 0ns
^? 2610:20:6f15:15::27 0 8 0 -+0ns [ +0ns]+/- 0ns
What is this source?
Post by Brian Gieryk
I don't see anything unusual(?) in the log...
Do you see anything? Why not give us 10 or 20 lines from the refclock log?
Post by Brian Gieryk
I am not an old hand at this......
Thank you!
Brian
Post by Bill Unruh
You tell it to lock to NMEA. Do you actually have an NMEA refclock running?
Ie you might want to show us what your complete chrony.conf is.
Also in your sources, what sources are listed? Ie, what is the complete output of chronyc sources command.
In the logging of the refclock, are there any entries? Do they give you a
clue?
William G. Unruh __| Canadian Institute for|____ Tel: +1(604)822-3273
Physics&Astronomy _|___ Advanced Research _|____ Fax: +1(604)822-5324
Canada V6T 1Z1 ____|____ and Gravity ______|_ www.theory.physics.ubc.ca/
Post by Brian Gieryk
Good morning.
Running chrony 3.1, raspberry Pi 3 model B with the latest Raspbian Jessie.
Have GPS working to set the clock with no issue.
Have good PPS on /dev/pps0
refclock PPS /dev/pps0 poll 0 refid PPS lock NMEA
Appended to the end of /etc/chrony/chrony.conf
Running chronyd produces no errors
PPS shows in chronyc sources
However the #? next to PPS never changes (48 hours so far) and there are no corrections.
Any thoughts?
Thank you all!
Brian
Sent from my iPhone
--
with "unsubscribe" in the subject.
with "help" in the subject.
--
--
with "unsubscribe" in the subject.
with "help" in the subject.
--
--
with "unsubscribe" in the subject.
with "help" in the subject.
--
--
To unsubscribe email chrony-users-***@chrony.tuxfamily.org
with "unsubscribe" in the subject.
For help email chrony-users-***@chrony.tuxfamily.org
with "help" in the subject.
Trouble? Email ***@chrony.tuxfamily.org.
Miroslav Lichvar
2017-02-24 15:36:37 UTC
Permalink
Post by Brian Gieryk
Ok, there is one GPS (with PPS out on a second output, going to GPIO 18 on the rpi)
I did a little change (commented out the two NMEA sources.
I am running gpsd for the GPS source.
Ok, the PPS input is on a GPIO instead of serial port. That makes a
difference. gpsd probably doesn't know that /dev/pps0 belongs to the
GPS and it doesn't process the PPS signal. Adding /dev/pps0 to the
gpsd command line in addition to the serial device might help.

If that doesn't work, you can use the PPS device directly with the PPS
refclock and use gpsd just for the message timing data. Like this:

refclock SHM 0 refid NMEA noselect
refclock PPS /dev/pps0 lock NMEA refid GPS
--
Miroslav Lichvar
--
To unsubscribe email chrony-users-***@chrony.tuxfamily.org
with "unsubscribe" in the subject.
For help email chrony-users-***@chrony.tuxfamily.org
with "help" in the subject.
Trouble? Email ***@chrony.tuxfamily.org.
Bill Unruh
2017-02-24 15:40:45 UTC
Permalink
William G. Unruh __| Canadian Institute for|____ Tel: +1(604)822-3273
Physics&Astronomy _|___ Advanced Research _|____ Fax: +1(604)822-5324
UBC, Vancouver,BC _|_ Program in Cosmology |____ ***@physics.ubc.ca
Canada V6T 1Z1 ____|____ and Gravity ______|_ www.theory.physics.ubc.ca/
Post by Brian Gieryk
Ok, there is one GPS (with PPS out on a second output, going to GPIO 18 on the rpi)
I did a little change (commented out the two NMEA sources.
I am running gpsd for the GPS source.
It is a Dragino LoRaWAN/GPS hat.
As I mentioned, the GPS is functioning (what appears to me) correctly.
I will add refclocks logging right now.
I will most likely add a second GPS, or more probably, tie it into my Rubidium standard.
As for the PPS signal, ppstest /dev/pps0 give the expected output, finds the PPS signal, and is happy with it.
Chronyc sources just does not seem to recognize it.
No, the problem probably is that PPS only gives time to the second and gives
no idea which second it is. Ie, it is time%1sec. It needs something else and
it needs to know that that something else is actually already working to give
the time with an accuaracy of a bit better than a second. In your case you
told the pps to use nmea, and nmea was not working at all, so the PPS had
nothing to use to tell it which second it was.
Lichvar gave you some suggestions. You might try them.
Post by Brian Gieryk
Thank you!
Brian
Sent from my iPhone
Post by Bill Unruh
Post by Brian Gieryk
I apologize for my seeming ignorance, I have been trying everything I could find in the archives here.
No problem. JUst remember that we have no information about your system except
what you tell us.
Post by Brian Gieryk
I have been using Linux for a grand total of one week, so some patience will be much appreciated.
My Pi does not seem to send email directly to anyone other than myself, so it is a 2 step process to send anything from it.
As for locking to the GPS, the time/date are being set correctly from it now, I would just like some redundancy, and a tad more accuracy, if it can be had.
But I am a bit confused. Do you have two or 3 GPS receivers attached to
your Pi? Or are all of GPS, the two NMEA and PPS coming (as I suspect) from one GPS receiver. If my suspicion is true, then that is not redundancy.
The various attempts t read 3 different times from the same receiver could
actually make things worse, not better (the readings fighting with each other).
Secondly try to get one thing going at a time. GPS, whatever it is (what
reports the gps to the SHM?-- are you using gpsd?) seems to work now. THe NMEA
are not. Use what works to gine the PPS its seconds-resolution, so PPS can
deliver its soperior accuracy, and so you can see if you can get it working.
Worry about redundancy later.
Post by Brian Gieryk
I will get more info shortly.
Thank you for your understanding.
Brian
Sent from my iPhone
Post by Bill Unruh
You do not know what is relevant do you?
Post by Brian Gieryk
# offline sources
refclock SHM 0 refid GPS precision 1e-1
What is putting this info into shm?
Post by Brian Gieryk
refclock SHM 1 refid NMEA precision 1e-9
refclock SOCK /tmp/chrony.ttyS0.sock refid NMEA
Two different sources with the same name. That could cause confusion. Where do these come from?
Post by Brian Gieryk
refclock PPS /dev/pps0 poll 0 refid PPS lock NMEA
210 Number of sources = 5
MS Name/IP address Stratum Poll Reach LastRx Last
sample
===============================================================================
#* GPS 0 4 377 15-+1007us [-1476us]+/-100ms
GPS reports. So why not lock to it?
Post by Brian Gieryk
#? NMEA 0 4 0 -+0ns [ +0ns]+/- 0ns
#? NMEA 0 4 0 -+0ns [ +0ns]+/- 0ns
Neither of your nmea sources report so PPS has nothing to lock to.
Post by Brian Gieryk
#? PPS 0 0 0 -+0ns [ +0ns]+/- 0ns
^? 2610:20:6f15:15::27 0 8 0 -+0ns [ +0ns]+/- 0ns
What is this source?
Post by Brian Gieryk
I don't see anything unusual(?) in the log...
Do you see anything? Why not give us 10 or 20 lines from the refclock log?
Post by Brian Gieryk
I am not an old hand at this......
Thank you!
Brian
Post by Bill Unruh
You tell it to lock to NMEA. Do you actually have an NMEA refclock running?
Ie you might want to show us what your complete chrony.conf is.
Also in your sources, what sources are listed? Ie, what is the complete output of chronyc sources command.
In the logging of the refclock, are there any entries? Do they give you a
clue?
William G. Unruh __| Canadian Institute for|____ Tel: +1(604)822-3273
Physics&Astronomy _|___ Advanced Research _|____ Fax: +1(604)822-5324
Canada V6T 1Z1 ____|____ and Gravity ______|_ www.theory.physics.ubc.ca/
Post by Brian Gieryk
Good morning.
Running chrony 3.1, raspberry Pi 3 model B with the latest Raspbian Jessie.
Have GPS working to set the clock with no issue.
Have good PPS on /dev/pps0
refclock PPS /dev/pps0 poll 0 refid PPS lock NMEA
Appended to the end of /etc/chrony/chrony.conf
Running chronyd produces no errors
PPS shows in chronyc sources
However the #? next to PPS never changes (48 hours so far) and there are no corrections.
Any thoughts?
Thank you all!
Brian
Sent from my iPhone
--
with "unsubscribe" in the subject.
with "help" in the subject.
--
--
with "unsubscribe" in the subject.
with "help" in the subject.
--
--
with "unsubscribe" in the subject.
with "help" in the subject.
--
--
with "unsubscribe" in the subject.
with "help" in the subject.
--
To unsubscribe email chrony-users-***@chrony.tuxfamily.org
with "unsubscribe" in the subject.
For help email chrony-users-***@chrony.tuxfamily.org
with "help" in the subject.
Trouble? Email ***@chrony.tuxfamily.org.
Brian Gieryk
2017-02-24 15:44:02 UTC
Permalink
Did as suggested, and seems to be working. NMEA is showing +/- 2599 us and GPS (PPS) is showing +/- 1831 us.

Thank you both!

Brian

Sent from my iPhone
Post by Bill Unruh
William G. Unruh __| Canadian Institute for|____ Tel: +1(604)822-3273
Physics&Astronomy _|___ Advanced Research _|____ Fax: +1(604)822-5324
Canada V6T 1Z1 ____|____ and Gravity ______|_ www.theory.physics.ubc.ca/
Post by Brian Gieryk
Ok, there is one GPS (with PPS out on a second output, going to GPIO 18 on the rpi)
I did a little change (commented out the two NMEA sources.
I am running gpsd for the GPS source.
It is a Dragino LoRaWAN/GPS hat.
As I mentioned, the GPS is functioning (what appears to me) correctly.
I will add refclocks logging right now.
I will most likely add a second GPS, or more probably, tie it into my Rubidium standard.
As for the PPS signal, ppstest /dev/pps0 give the expected output, finds the PPS signal, and is happy with it.
Chronyc sources just does not seem to recognize it.
No, the problem probably is that PPS only gives time to the second and gives
no idea which second it is. Ie, it is time%1sec. It needs something else and
it needs to know that that something else is actually already working to give
the time with an accuaracy of a bit better than a second. In your case you
told the pps to use nmea, and nmea was not working at all, so the PPS had
nothing to use to tell it which second it was. Lichvar gave you some suggestions. You might try them.
Post by Brian Gieryk
Thank you!
Brian
Sent from my iPhone
Post by Bill Unruh
Post by Brian Gieryk
I apologize for my seeming ignorance, I have been trying everything I could find in the archives here.
No problem. JUst remember that we have no information about your system except
what you tell us.
Post by Brian Gieryk
I have been using Linux for a grand total of one week, so some patience will be much appreciated.
My Pi does not seem to send email directly to anyone other than myself, so it is a 2 step process to send anything from it.
As for locking to the GPS, the time/date are being set correctly from it now, I would just like some redundancy, and a tad more accuracy, if it can be had.
But I am a bit confused. Do you have two or 3 GPS receivers attached to
your Pi? Or are all of GPS, the two NMEA and PPS coming (as I suspect) from one GPS receiver. If my suspicion is true, then that is not redundancy.
The various attempts t read 3 different times from the same receiver could
actually make things worse, not better (the readings fighting with each other).
Secondly try to get one thing going at a time. GPS, whatever it is (what
reports the gps to the SHM?-- are you using gpsd?) seems to work now. THe NMEA
are not. Use what works to gine the PPS its seconds-resolution, so PPS can
deliver its soperior accuracy, and so you can see if you can get it working.
Worry about redundancy later.
Post by Brian Gieryk
I will get more info shortly.
Thank you for your understanding.
Brian
Sent from my iPhone
Post by Bill Unruh
You do not know what is relevant do you?
Post by Brian Gieryk
# offline sources
refclock SHM 0 refid GPS precision 1e-1
What is putting this info into shm?
Post by Brian Gieryk
refclock SHM 1 refid NMEA precision 1e-9
refclock SOCK /tmp/chrony.ttyS0.sock refid NMEA
Two different sources with the same name. That could cause confusion. Where do these come from?
Post by Brian Gieryk
refclock PPS /dev/pps0 poll 0 refid PPS lock NMEA
210 Number of sources = 5
MS Name/IP address Stratum Poll Reach LastRx Last
sample
===============================================================================
#* GPS 0 4 377 15-+1007us [-1476us]+/-100ms
GPS reports. So why not lock to it?
Post by Brian Gieryk
#? NMEA 0 4 0 -+0ns [ +0ns]+/- 0ns
#? NMEA 0 4 0 -+0ns [ +0ns]+/- 0ns
Neither of your nmea sources report so PPS has nothing to lock to.
Post by Brian Gieryk
#? PPS 0 0 0 -+0ns [ +0ns]+/- 0ns
^? 2610:20:6f15:15::27 0 8 0 -+0ns [ +0ns]+/- 0ns
What is this source?
Post by Brian Gieryk
I don't see anything unusual(?) in the log...
Do you see anything? Why not give us 10 or 20 lines from the refclock log?
Post by Brian Gieryk
I am not an old hand at this......
Thank you!
Brian
Post by Bill Unruh
You tell it to lock to NMEA. Do you actually have an NMEA refclock running?
Ie you might want to show us what your complete chrony.conf is.
Also in your sources, what sources are listed? Ie, what is the complete output of chronyc sources command.
In the logging of the refclock, are there any entries? Do they give you a
clue?
William G. Unruh __| Canadian Institute for|____ Tel: +1(604)822-3273
Physics&Astronomy _|___ Advanced Research _|____ Fax: +1(604)822-5324
Canada V6T 1Z1 ____|____ and Gravity ______|_ www.theory.physics.ubc.ca/
Post by Brian Gieryk
Good morning.
Running chrony 3.1, raspberry Pi 3 model B with the latest Raspbian Jessie.
Have GPS working to set the clock with no issue.
Have good PPS on /dev/pps0
refclock PPS /dev/pps0 poll 0 refid PPS lock NMEA
Appended to the end of /etc/chrony/chrony.conf
Running chronyd produces no errors
PPS shows in chronyc sources
However the #? next to PPS never changes (48 hours so far) and there are no corrections.
Any thoughts?
Thank you all!
Brian
Sent from my iPhone
--
with "unsubscribe" in the subject.
with "help" in the subject.
--
--
with "unsubscribe" in the subject.
with "help" in the subject.
--
--
with "unsubscribe" in the subject.
with "help" in the subject.
--
--
with "unsubscribe" in the subject.
with "help" in the subject.
--
--
To unsubscribe email chrony-users-***@chrony.tuxfamily.org
with "unsubscribe" in the subject.
For help email chrony-users-***@chrony.tuxfamily.org
with "help" in the subject.
Trouble? Email ***@chrony.tuxfamily.org.
Miroslav Lichvar
2017-02-24 15:48:28 UTC
Permalink
Post by Brian Gieryk
Did as suggested, and seems to be working. NMEA is showing +/- 2599 us and GPS (PPS) is showing +/- 1831 us.
1800 microseconds is not that good for a PPS. Usually it's just a few
microseconds, or better. What does the "chronyc sourcestats" command
print?
--
Miroslav Lichvar
--
To unsubscribe email chrony-users-***@chrony.tuxfamily.org
with "unsubscribe" in the subject.
For help email chrony-users-***@chrony.tuxfamily.org
with "help" in the subject.
Trouble? Email ***@chrony.tuxfamily.org.
Brian Gieryk
2017-02-24 15:53:12 UTC
Permalink
Bill, PPS is being referenced as GPS now.

Miroslav, that was initial fix.

Now, NMEA offset is +198ms std dev is 7301us

GPS (PPS) is offset +23 us std dev is 19us.

Brian

Sent from my iPhone
Post by Miroslav Lichvar
Post by Brian Gieryk
Did as suggested, and seems to be working. NMEA is showing +/- 2599 us and GPS (PPS) is showing +/- 1831 us.
1800 microseconds is not that good for a PPS. Usually it's just a few
microseconds, or better. What does the "chronyc sourcestats" command
print?
--
Miroslav Lichvar
--
with "unsubscribe" in the subject.
with "help" in the subject.
--
To unsubscribe email chrony-users-***@chrony.tuxfamily.org
with "unsubscribe" in the subject.
For help email chrony-users-***@chrony.tuxfamily.org
with "help" in the subject.
Trouble? Email ***@chrony.tuxfamily.org.
Fabian Schörghofer
2017-02-24 15:57:15 UTC
Permalink
What does your chrony.conf look like now?

I have a similar problem, (also using gpsd and Raspberry Pi with GPIO-PPS).
Post by Brian Gieryk
Bill, PPS is being referenced as GPS now.
Miroslav, that was initial fix.
Now, NMEA offset is +198ms std dev is 7301us
GPS (PPS) is offset +23 us std dev is 19us.
Brian
Sent from my iPhone
Post by Miroslav Lichvar
Post by Brian Gieryk
Did as suggested, and seems to be working. NMEA is showing +/- 2599 us and GPS (PPS) is showing +/- 1831 us.
1800 microseconds is not that good for a PPS. Usually it's just a few
microseconds, or better. What does the "chronyc sourcestats" command
print?
--
Miroslav Lichvar
--
with "unsubscribe" in the subject.
with "help" in the subject.
--
To unsubscribe email chrony-users-***@chrony.tuxfamily.org
with "unsubscribe" in the subject.
For help email chrony-users-***@chrony.tuxfamily.org
with "help" in the subject.
Trouble? Email ***@chrony.tuxfamily.org.
Brian Gieryk
2017-02-24 16:02:51 UTC
Permalink
This is mine, most online sources are commented out, as internet is
very limited here.

Brian

On Fri, 24 Feb 2017 16:57:15 +0100
Post by Fabian Schörghofer
What does your chrony.conf look like now?
I have a similar problem, (also using gpsd and Raspberry Pi with GPIO-PPS).
Post by Brian Gieryk
Bill, PPS is being referenced as GPS now.
Miroslav, that was initial fix.
Now, NMEA offset is +198ms std dev is 7301us
GPS (PPS) is offset +23 us std dev is 19us.
Brian
Sent from my iPhone
Post by Miroslav Lichvar
Post by Brian Gieryk
Did as suggested, and seems to be working. NMEA is showing +/-
2599 us and GPS (PPS) is showing +/- 1831 us.
1800 microseconds is not that good for a PPS. Usually it's just a
few microseconds, or better. What does the "chronyc sourcestats"
command print?
--
Miroslav Lichvar
--
with "unsubscribe" in the subject.
with "help" in the subject.
Bill Unruh
2017-02-24 18:28:05 UTC
Permalink
Post by Brian Gieryk
Bill, PPS is being referenced as GPS now.
Miroslav, that was initial fix.
Now, NMEA offset is +198ms std dev is 7301us
Sounds about right.
Post by Brian Gieryk
GPS (PPS) is offset +23 us std dev is 19us.
Seems a bit high. But I do not know either your gps or how well the PI
interrupt works.
Post by Brian Gieryk
Brian
Sent from my iPhone
Post by Miroslav Lichvar
Post by Brian Gieryk
Did as suggested, and seems to be working. NMEA is showing +/- 2599 us and GPS (PPS) is showing +/- 1831 us.
1800 microseconds is not that good for a PPS. Usually it's just a few
microseconds, or better. What does the "chronyc sourcestats" command
print?
--
Miroslav Lichvar
--
with "unsubscribe" in the subject.
with "help" in the subject.
--
with "unsubscribe" in the subject.
with "help" in the subject.
--
To unsubscribe email chrony-users-***@chrony.tuxfamily.org
with "unsubscribe" in the subject.
For help email chrony-users-***@chrony.tuxfamily.org
with "help" in the subject.
Trouble? Email ***@chrony.tuxfamily.org.
Bill Unruh
2017-02-24 08:00:42 UTC
Permalink
You are not logging a crucial thing, refclocks logging.
Put refclocks into the logging line in chrony.conf


Note-- it is pps that gines the accuracy. And you need to lock PPS to a
working other source of inaccurate time (better than seconds)



William G. Unruh __| Canadian Institute for|____ Tel: +1(604)822-3273
Physics&Astronomy _|___ Advanced Research _|____ Fax: +1(604)822-5324
UBC, Vancouver,BC _|_ Program in Cosmology |____ ***@physics.ubc.ca
Canada V6T 1Z1 ____|____ and Gravity ______|_ www.theory.physics.ubc.ca/
Post by Brian Gieryk
I apologize for my seeming ignorance, I have been trying everything I could find in the archives here.
I have been using Linux for a grand total of one week, so some patience will be much appreciated.
My Pi does not seem to send email directly to anyone other than myself, so it is a 2 step process to send anything from it.
As for locking to the GPS, the time/date are being set correctly from it now, I would just like some redundancy, and a tad more accuracy, if it can be had.
I will get more info shortly.
Thank you for your understanding.
Brian
Sent from my iPhone
Post by Bill Unruh
You do not know what is relevant do you?
Post by Brian Gieryk
# offline sources
refclock SHM 0 refid GPS precision 1e-1
What is putting this info into shm?
Post by Brian Gieryk
refclock SHM 1 refid NMEA precision 1e-9
refclock SOCK /tmp/chrony.ttyS0.sock refid NMEA
Two different sources with the same name. That could cause confusion. Where do these come from?
Post by Brian Gieryk
refclock PPS /dev/pps0 poll 0 refid PPS lock NMEA
210 Number of sources = 5
MS Name/IP address Stratum Poll Reach LastRx Last
sample
===============================================================================
#* GPS 0 4 377 15-+1007us [-1476us]+/-100ms
GPS reports. So why not lock to it?
Post by Brian Gieryk
#? NMEA 0 4 0 -+0ns [ +0ns]+/- 0ns
#? NMEA 0 4 0 -+0ns [ +0ns]+/- 0ns
Neither of your nmea sources report so PPS has nothing to lock to.
Post by Brian Gieryk
#? PPS 0 0 0 -+0ns [ +0ns]+/- 0ns
^? 2610:20:6f15:15::27 0 8 0 -+0ns [ +0ns]+/- 0ns
What is this source?
Post by Brian Gieryk
I don't see anything unusual(?) in the log...
Do you see anything? Why not give us 10 or 20 lines from the refclock log?
Post by Brian Gieryk
I am not an old hand at this......
Thank you!
Brian
Post by Bill Unruh
You tell it to lock to NMEA. Do you actually have an NMEA refclock running?
Ie you might want to show us what your complete chrony.conf is.
Also in your sources, what sources are listed? Ie, what is the complete output of chronyc sources command.
In the logging of the refclock, are there any entries? Do they give you a
clue?
William G. Unruh __| Canadian Institute for|____ Tel: +1(604)822-3273
Physics&Astronomy _|___ Advanced Research _|____ Fax: +1(604)822-5324
Canada V6T 1Z1 ____|____ and Gravity ______|_ www.theory.physics.ubc.ca/
Post by Brian Gieryk
Good morning.
Running chrony 3.1, raspberry Pi 3 model B with the latest Raspbian Jessie.
Have GPS working to set the clock with no issue.
Have good PPS on /dev/pps0
refclock PPS /dev/pps0 poll 0 refid PPS lock NMEA
Appended to the end of /etc/chrony/chrony.conf
Running chronyd produces no errors
PPS shows in chronyc sources
However the #? next to PPS never changes (48 hours so far) and there are no corrections.
Any thoughts?
Thank you all!
Brian
Sent from my iPhone
--
with "unsubscribe" in the subject.
with "help" in the subject.
--
--
with "unsubscribe" in the subject.
with "help" in the subject.
--
--
with "unsubscribe" in the subject.
with "help" in the subject.
--
To unsubscribe email chrony-users-***@chrony.tuxfamily.org
with "unsubscribe" in the subject.
For help email chrony-users-***@chrony.tuxfamily.org
with "help" in the subject.
Trouble? Email ***@chrony.tuxfamily.org.
Miroslav Lichvar
2017-02-24 08:34:13 UTC
Permalink
Post by Bill Unruh
You are not logging a crucial thing, refclocks logging.
Put refclocks into the logging line in chrony.conf
Note-- it is pps that gines the accuracy. And you need to lock PPS to a
working other source of inaccurate time (better than seconds)
It would be also good to know what GPS receiver is it. Maybe it
doesn't have PPS or it is not working? gpsd prints some useful
messages in the debug mode. E.g.

gpsd -n -N /dev/ttyS0 -D 9 2>&1 | grep PPS

Anyway, there are several options when configuring chrony to use GPS
with gpsd. I think the best options are:

A) refclock SHM 1

B) refclock SOCK /var/run/chrony.ttyS0.sock

C) refclock SHM 0 refid NMEA noselect
refclock PPS /dev/pps0 lock NMEA

Only one should be used at a time. B requires chronyd to be started
before gpsd. C requires gpsd (or ldattach) to be started before
chronyd. In A the order doesn't matter.
--
Miroslav Lichvar
--
To unsubscribe email chrony-users-***@chrony.tuxfamily.org
with "unsubscribe" in the subject.
For help email chrony-users-***@chrony.tuxfamily.org
with "help" in the subject.
Trouble? Email ***@chrony.tuxfamily.org.
Brian Gieryk
2017-02-24 15:22:49 UTC
Permalink
Recent snipet from refclocks.log


===============================================================================
Date (UTC) Time Refid DP L P Raw offset Cooked offset Disp.
===============================================================================
2017-02-24 15:20:01.987157 GPS 10 N 0 1.375465e-02 1.284218e-02 1.000e-01
2017-02-24 15:20:03.007568 GPS 11 N 0 -6.695705e-03 -7.568935e-03 1.000e-01
2017-02-24 15:20:04.006286 GPS 12 N 0 -5.451906e-03 -6.286731e-03 1.000e-01
2017-02-24 15:20:04.987341 GPS 13 N 0 1.345536e-02 1.265826e-02 1.000e-01
2017-02-24 15:20:05.981351 GPS 14 N 0 1.940777e-02 1.864889e-02 1.000e-01
2017-02-24 15:20:07.017450 GPS 15 N 0 -1.673142e-02 -1.745045e-02 1.000e-01
2017-02-24 15:19:58.797466 GPS - N - - 2.533224e-03 1.000e-01
2017-02-24 15:20:08.004129 GPS 0 N 0 -1.242240e-03 -4.129600e-03 1.000e-01
2017-02-24 15:20:09.003884 GPS 1 N 0 -1.233461e-03 -3.884212e-03 1.000e-01
2017-02-24 15:20:10.016937 GPS 2 N 0 -1.452626e-02 -1.693726e-02 1.000e-01
2017-02-24 15:20:10.991004 GPS 3 N 0 1.117600e-02 8.995530e-03 1.000e-01
2017-02-24 15:20:11.990037 GPS 4 N 0 1.190633e-02 9.962303e-03 1.000e-01
2017-02-24 15:20:13.020436 GPS 5 N 0 -1.873611e-02 -2.043628e-02 1.000e-01
2017-02-24 15:20:13.990162 GPS 6 N 0 1.130804e-02 9.837376e-03 1.000e-01
2017-02-24 15:20:15.010200 GPS 7 N 0 -8.971401e-03 -1.020066e-02 1.000e-01
2017-02-24 15:20:15.984244 GPS 8 N 0 1.675461e-02 1.575588e-02 1.000e-01
2017-02-24 15:20:17.004312 GPS 9 N 0 -3.555564e-03 -4.312879e-03 1.000e-01
2017-02-24 15:20:18.004689 GPS 10 N 0 -4.168469e-03 -4.689028e-03 1.000e-01
2017-02-24 15:20:19.020420 GPS 11 N 0 -2.014009e-02 -2.042026e-02 1.000e-01
2017-02-24 15:20:20.010434 GPS 12 N 0 -1.043449e-02 -1.043431e-02 1.000e-01
2017-02-24 15:20:21.016526 GPS 13 N 0 -1.652631e-02 -1.652613e-02 1.000e-01
2017-02-24 15:20:21.984592 GPS 14 N 0 1.540710e-02 1.540728e-02 1.000e-01
2017-02-24 15:20:23.005002 GPS 15 N 0 -5.002770e-03 -5.002588e-03 1.000e-01
2017-02-24 15:20:15.604034 GPS - N - - -4.034649e-03 1.000e-01
2017-02-24 15:20:24.008004 GPS 0 N 0 -9.683123e-03 -8.004299e-03 1.000e-01
2017-02-24 15:20:25.025063 GPS 1 N 0 -2.666452e-02 -2.506397e-02 1.000e-01
2017-02-24 15:20:26.015081 GPS 2 N 0 -1.660632e-02 -1.508197e-02 1.000e-01
2017-02-24 15:20:27.015129 GPS 3 N 0 -1.657705e-02 -1.512967e-02 1.000e-01
2017-02-24 15:20:27.983524 GPS 4 N 0 1.510218e-02 1.647503e-02 1.000e-01
2017-02-24 15:20:28.988274 GPS 5 N 0 1.043042e-02 1.172593e-02 1.000e-01
2017-02-24 15:20:29.983295 GPS 6 N 0 1.548592e-02 1.670486e-02 1.000e-01
2017-02-24 15:20:31.000392 GPS 7 N 0 -1.533493e-03 -3.928390e-04 1.000e-01
===============================================================================
Date (UTC) Time Refid DP L P Raw offset Cooked offset Disp.
===============================================================================
2017-02-24 15:20:31.983413 GPS 8 N 0 1.552176e-02 1.658675e-02 1.000e-01
2017-02-24 15:20:32.962769 GPS 9 N 0 3.624121e-02 3.723083e-02 1.000e-01
2017-02-24 15:20:33.988544 GPS 10 N 0 1.054480e-02 1.145547e-02 1.000e-01
2017-02-24 15:20:34.988592 GPS 11 N 0 1.057412e-02 1.140782e-02 1.000e-01
2017-02-24 15:20:35.988681 GPS 12 N 0 1.056225e-02 1.131898e-02 1.000e-01
2017-02-24 15:20:36.998711 GPS 13 N 0 6.091520e-04 1.288145e-03 1.000e-01
2017-02-24 15:20:37.964039 GPS 14 N 0 3.535536e-02 3.596006e-02 1.000e-01
2017-02-24 15:20:38.988797 GPS 15 N 0 1.067647e-02 1.120229e-02 1.000e-01
2017-02-24 15:20:32.491693 GPS - N - - 8.306328e-03 1.000e-01
2017-02-24 15:20:39.994004 GPS 0 N 0 1.046571e-02 5.995506e-03 1.000e-01
2017-02-24 15:20:40.994121 GPS 1 N 0 9.877895e-03 5.878899e-03 1.000e-01
2017-02-24 15:20:41.994145 GPS 2 N 0 9.382104e-03 5.854275e-03 1.000e-01
2017-02-24 15:20:43.003612 GPS 3 N 0 -5.606230e-04 -3.612836e-03 1.000e-01
2017-02-24 15:20:43.993340 GPS 4 N 0 9.245389e-03 6.659491e-03 1.000e-01
2017-02-24 15:20:44.979309 GPS 5 N 0 2.281177e-02 2.069042e-02 1.000e-01
2017-02-24 15:20:45.979376 GPS 6 N 0 2.227403e-02 2.062386e-02 1.000e-01
2017-02-24 15:20:46.979469 GPS 7 N 0 2.170914e-02 2.053017e-02 1.000e-01
2017-02-24 15:20:47.975818 GPS 8 N 0 2.489076e-02 2.418123e-02 1.000e-01
2017-02-24 15:20:48.998301 GPS 9 N 0 1.697528e-03 1.698060e-03 1.000e-01
2017-02-24 15:20:49.968538 GPS 10 N 0 3.146090e-02 3.146143e-02 1.000e-01
2017-02-24 15:20:50.988606 GPS 11 N 0 1.139313e-02 1.139366e-02 1.000e-01
2017-02-24 15:20:51.989674 GPS 12 N 0 1.032455e-02 1.032508e-02 1.000e-01
2017-02-24 15:20:52.995053 GPS 13 N 0 4.946263e-03 4.946795e-03 1.000e-01
2017-02-24 15:20:53.994809 GPS 14 N 0 5.190453e-03 5.190985e-03 1.000e-01
2017-02-24 15:20:54.998813 GPS 15 N 0 1.185920e-03 1.186452e-03 1.000e-01
2017-02-24 15:20:46.990260 GPS - N - - 9.739873e-03 1.000e-01
2017-02-24 15:20:55.983874 GPS 0 N 0 2.454278e-02 1.612560e-02 1.000e-01
2017-02-24 15:20:57.002943 GPS 1 N 0 4.242812e-03 -2.943432e-03 1.000e-01
2017-02-24 15:20:58.003534 GPS 2 N 0 2.442725e-03 -3.534903e-03 1.000e-01
2017-02-24 15:20:58.998015 GPS 3 N 0 6.760802e-03 1.984408e-03 1.000e-01
2017-02-24 15:20:59.998056 GPS 4 N 0 5.511664e-03 1.943221e-03 1.000e-01
2017-02-24 15:21:01.013240 GPS 5 N 0 -1.089818e-02 -1.324038e-02 1.000e-01
===============================================================================
Date (UTC) Time Refid DP L P Raw offset Cooked offset Disp.
===============================================================================
2017-02-24 15:21:01.989195 GPS 6 N 0 1.196806e-02 1.080472e-02 1.000e-01
2017-02-24 15:21:03.009589 GPS 7 N 0 -9.589934e-03 -9.589271e-03 1.000e-01
2017-02-24 15:21:04.009321 GPS 8 N 0 -9.322357e-03 -9.321694e-03 1.000e-01
2017-02-24 15:21:04.983406 GPS 9 N 0 1.659283e-02 1.659349e-02 1.000e-01
2017-02-24 15:21:06.008378 GPS 10 N 0 -8.378905e-03 -8.378242e-03 1.000e-01
2017-02-24 15:21:07.019457 GPS 11 N 0 -1.945806e-02 -1.945740e-02
1.000e-01


Brian



Sent from my iPhone
Post by Miroslav Lichvar
Post by Bill Unruh
You are not logging a crucial thing, refclocks logging.
Put refclocks into the logging line in chrony.conf
Note-- it is pps that gines the accuracy. And you need to lock PPS to a
working other source of inaccurate time (better than seconds)
It would be also good to know what GPS receiver is it. Maybe it
doesn't have PPS or it is not working? gpsd prints some useful
messages in the debug mode. E.g.
gpsd -n -N /dev/ttyS0 -D 9 2>&1 | grep PPS
Anyway, there are several options when configuring chrony to use GPS
A) refclock SHM 1
B) refclock SOCK /var/run/chrony.ttyS0.sock
C) refclock SHM 0 refid NMEA noselect
refclock PPS /dev/pps0 lock NMEA
Only one should be used at a time. B requires chronyd to be started
before gpsd. C requires gpsd (or ldattach) to be started before
chronyd. In A the order doesn't matter.
--
Miroslav Lichvar
--
with "unsubscribe" in the subject.
with "help" in the subject.
--
To unsubscribe email chrony-users-***@chrony.tuxfamily.org
with "unsubscribe" in the subject.
For help email chrony-users-***@chrony.tuxfamily.org
with "help" in the subject.
Trouble? Email ***@chrony.tuxfamily.org.
Bill Unruh
2017-02-24 15:46:05 UTC
Permalink
Hm. I do not see the PPS refclock in there.



William G. Unruh __| Canadian Institute for|____ Tel: +1(604)822-3273
Physics&Astronomy _|___ Advanced Research _|____ Fax: +1(604)822-5324
UBC, Vancouver,BC _|_ Program in Cosmology |____ ***@physics.ubc.ca
Canada V6T 1Z1 ____|____ and Gravity ______|_ www.theory.physics.ubc.ca/
Post by Brian Gieryk
Recent snipet from refclocks.log
===============================================================================
Date (UTC) Time Refid DP L P Raw offset Cooked offset Disp.
===============================================================================
2017-02-24 15:20:01.987157 GPS 10 N 0 1.375465e-02 1.284218e-02 1.000e-01
2017-02-24 15:20:03.007568 GPS 11 N 0 -6.695705e-03 -7.568935e-03 1.000e-01
2017-02-24 15:20:04.006286 GPS 12 N 0 -5.451906e-03 -6.286731e-03 1.000e-01
2017-02-24 15:20:04.987341 GPS 13 N 0 1.345536e-02 1.265826e-02 1.000e-01
2017-02-24 15:20:05.981351 GPS 14 N 0 1.940777e-02 1.864889e-02 1.000e-01
2017-02-24 15:20:07.017450 GPS 15 N 0 -1.673142e-02 -1.745045e-02 1.000e-01
2017-02-24 15:19:58.797466 GPS - N - - 2.533224e-03 1.000e-01
2017-02-24 15:20:08.004129 GPS 0 N 0 -1.242240e-03 -4.129600e-03 1.000e-01
2017-02-24 15:20:09.003884 GPS 1 N 0 -1.233461e-03 -3.884212e-03 1.000e-01
2017-02-24 15:20:10.016937 GPS 2 N 0 -1.452626e-02 -1.693726e-02 1.000e-01
2017-02-24 15:20:10.991004 GPS 3 N 0 1.117600e-02 8.995530e-03 1.000e-01
2017-02-24 15:20:11.990037 GPS 4 N 0 1.190633e-02 9.962303e-03 1.000e-01
2017-02-24 15:20:13.020436 GPS 5 N 0 -1.873611e-02 -2.043628e-02 1.000e-01
2017-02-24 15:20:13.990162 GPS 6 N 0 1.130804e-02 9.837376e-03 1.000e-01
2017-02-24 15:20:15.010200 GPS 7 N 0 -8.971401e-03 -1.020066e-02 1.000e-01
2017-02-24 15:20:15.984244 GPS 8 N 0 1.675461e-02 1.575588e-02 1.000e-01
2017-02-24 15:20:17.004312 GPS 9 N 0 -3.555564e-03 -4.312879e-03 1.000e-01
2017-02-24 15:20:18.004689 GPS 10 N 0 -4.168469e-03 -4.689028e-03 1.000e-01
2017-02-24 15:20:19.020420 GPS 11 N 0 -2.014009e-02 -2.042026e-02 1.000e-01
2017-02-24 15:20:20.010434 GPS 12 N 0 -1.043449e-02 -1.043431e-02 1.000e-01
2017-02-24 15:20:21.016526 GPS 13 N 0 -1.652631e-02 -1.652613e-02 1.000e-01
2017-02-24 15:20:21.984592 GPS 14 N 0 1.540710e-02 1.540728e-02 1.000e-01
2017-02-24 15:20:23.005002 GPS 15 N 0 -5.002770e-03 -5.002588e-03 1.000e-01
2017-02-24 15:20:15.604034 GPS - N - - -4.034649e-03 1.000e-01
2017-02-24 15:20:24.008004 GPS 0 N 0 -9.683123e-03 -8.004299e-03 1.000e-01
2017-02-24 15:20:25.025063 GPS 1 N 0 -2.666452e-02 -2.506397e-02 1.000e-01
2017-02-24 15:20:26.015081 GPS 2 N 0 -1.660632e-02 -1.508197e-02 1.000e-01
2017-02-24 15:20:27.015129 GPS 3 N 0 -1.657705e-02 -1.512967e-02 1.000e-01
2017-02-24 15:20:27.983524 GPS 4 N 0 1.510218e-02 1.647503e-02 1.000e-01
2017-02-24 15:20:28.988274 GPS 5 N 0 1.043042e-02 1.172593e-02 1.000e-01
2017-02-24 15:20:29.983295 GPS 6 N 0 1.548592e-02 1.670486e-02 1.000e-01
2017-02-24 15:20:31.000392 GPS 7 N 0 -1.533493e-03 -3.928390e-04 1.000e-01
===============================================================================
Date (UTC) Time Refid DP L P Raw offset Cooked offset Disp.
===============================================================================
2017-02-24 15:20:31.983413 GPS 8 N 0 1.552176e-02 1.658675e-02 1.000e-01
2017-02-24 15:20:32.962769 GPS 9 N 0 3.624121e-02 3.723083e-02 1.000e-01
2017-02-24 15:20:33.988544 GPS 10 N 0 1.054480e-02 1.145547e-02 1.000e-01
2017-02-24 15:20:34.988592 GPS 11 N 0 1.057412e-02 1.140782e-02 1.000e-01
2017-02-24 15:20:35.988681 GPS 12 N 0 1.056225e-02 1.131898e-02 1.000e-01
2017-02-24 15:20:36.998711 GPS 13 N 0 6.091520e-04 1.288145e-03 1.000e-01
2017-02-24 15:20:37.964039 GPS 14 N 0 3.535536e-02 3.596006e-02 1.000e-01
2017-02-24 15:20:38.988797 GPS 15 N 0 1.067647e-02 1.120229e-02 1.000e-01
2017-02-24 15:20:32.491693 GPS - N - - 8.306328e-03 1.000e-01
2017-02-24 15:20:39.994004 GPS 0 N 0 1.046571e-02 5.995506e-03 1.000e-01
2017-02-24 15:20:40.994121 GPS 1 N 0 9.877895e-03 5.878899e-03 1.000e-01
2017-02-24 15:20:41.994145 GPS 2 N 0 9.382104e-03 5.854275e-03 1.000e-01
2017-02-24 15:20:43.003612 GPS 3 N 0 -5.606230e-04 -3.612836e-03 1.000e-01
2017-02-24 15:20:43.993340 GPS 4 N 0 9.245389e-03 6.659491e-03 1.000e-01
2017-02-24 15:20:44.979309 GPS 5 N 0 2.281177e-02 2.069042e-02 1.000e-01
2017-02-24 15:20:45.979376 GPS 6 N 0 2.227403e-02 2.062386e-02 1.000e-01
2017-02-24 15:20:46.979469 GPS 7 N 0 2.170914e-02 2.053017e-02 1.000e-01
2017-02-24 15:20:47.975818 GPS 8 N 0 2.489076e-02 2.418123e-02 1.000e-01
2017-02-24 15:20:48.998301 GPS 9 N 0 1.697528e-03 1.698060e-03 1.000e-01
2017-02-24 15:20:49.968538 GPS 10 N 0 3.146090e-02 3.146143e-02 1.000e-01
2017-02-24 15:20:50.988606 GPS 11 N 0 1.139313e-02 1.139366e-02 1.000e-01
2017-02-24 15:20:51.989674 GPS 12 N 0 1.032455e-02 1.032508e-02 1.000e-01
2017-02-24 15:20:52.995053 GPS 13 N 0 4.946263e-03 4.946795e-03 1.000e-01
2017-02-24 15:20:53.994809 GPS 14 N 0 5.190453e-03 5.190985e-03 1.000e-01
2017-02-24 15:20:54.998813 GPS 15 N 0 1.185920e-03 1.186452e-03 1.000e-01
2017-02-24 15:20:46.990260 GPS - N - - 9.739873e-03 1.000e-01
2017-02-24 15:20:55.983874 GPS 0 N 0 2.454278e-02 1.612560e-02 1.000e-01
2017-02-24 15:20:57.002943 GPS 1 N 0 4.242812e-03 -2.943432e-03 1.000e-01
2017-02-24 15:20:58.003534 GPS 2 N 0 2.442725e-03 -3.534903e-03 1.000e-01
2017-02-24 15:20:58.998015 GPS 3 N 0 6.760802e-03 1.984408e-03 1.000e-01
2017-02-24 15:20:59.998056 GPS 4 N 0 5.511664e-03 1.943221e-03 1.000e-01
2017-02-24 15:21:01.013240 GPS 5 N 0 -1.089818e-02 -1.324038e-02 1.000e-01
===============================================================================
Date (UTC) Time Refid DP L P Raw offset Cooked offset Disp.
===============================================================================
2017-02-24 15:21:01.989195 GPS 6 N 0 1.196806e-02 1.080472e-02 1.000e-01
2017-02-24 15:21:03.009589 GPS 7 N 0 -9.589934e-03 -9.589271e-03 1.000e-01
2017-02-24 15:21:04.009321 GPS 8 N 0 -9.322357e-03 -9.321694e-03 1.000e-01
2017-02-24 15:21:04.983406 GPS 9 N 0 1.659283e-02 1.659349e-02 1.000e-01
2017-02-24 15:21:06.008378 GPS 10 N 0 -8.378905e-03 -8.378242e-03 1.000e-01
2017-02-24 15:21:07.019457 GPS 11 N 0 -1.945806e-02 -1.945740e-02
1.000e-01
Brian
ￌ
Sent from my iPhone
Post by Miroslav Lichvar
Post by Bill Unruh
You are not logging a crucial thing, refclocks logging.
Put refclocks into the logging line in chrony.conf
Note-- it is pps that gines the accuracy. And you need to lock PPS to a
working other source of inaccurate time (better than seconds)
It would be also good to know what GPS receiver is it. Maybe it
doesn't have PPS or it is not working? gpsd prints some useful
messages in the debug mode. E.g.
gpsd -n -N /dev/ttyS0 -D 9 2>&1 | grep PPS
Anyway, there are several options when configuring chrony to use GPS
A) refclock SHM 1
B) refclock SOCK /var/run/chrony.ttyS0.sock
C) refclock SHM 0 refid NMEA noselect
refclock PPS /dev/pps0 lock NMEA
Only one should be used at a time. B requires chronyd to be started
before gpsd. C requires gpsd (or ldattach) to be started before
chronyd. In A the order doesn't matter.
--
Miroslav Lichvar
--
with "unsubscribe" in the subject.
with "help" in the subject.
--
with "unsubscribe" in the subject.
with "help" in the subject.
Brian Gieryk
2017-02-24 15:32:22 UTC
Permalink
Smallish update:

I changed (in chrony.conf) GPS precision to 1e-6 and added PPS precision at 1e-9.

Now chronyc sources is seeing PPS!

I will let it cook awhile.

Thank you!

Brian

Sent from my iPhone
Post by Miroslav Lichvar
Post by Bill Unruh
You are not logging a crucial thing, refclocks logging.
Put refclocks into the logging line in chrony.conf
Note-- it is pps that gines the accuracy. And you need to lock PPS to a
working other source of inaccurate time (better than seconds)
It would be also good to know what GPS receiver is it. Maybe it
doesn't have PPS or it is not working? gpsd prints some useful
messages in the debug mode. E.g.
gpsd -n -N /dev/ttyS0 -D 9 2>&1 | grep PPS
Anyway, there are several options when configuring chrony to use GPS
A) refclock SHM 1
B) refclock SOCK /var/run/chrony.ttyS0.sock
C) refclock SHM 0 refid NMEA noselect
refclock PPS /dev/pps0 lock NMEA
Only one should be used at a time. B requires chronyd to be started
before gpsd. C requires gpsd (or ldattach) to be started before
chronyd. In A the order doesn't matter.
--
Miroslav Lichvar
--
with "unsubscribe" in the subject.
with "help" in the subject.
--
To unsubscribe email chrony-users-***@chrony.tuxfamily.org
with "unsubscribe" in the subject.
For help email chrony-users-***@chrony.tuxfamily.org
with "help" in the subject.
Trouble? Email ***@chrony.tuxfamily.org.
Brian Gieryk
2017-02-24 01:55:25 UTC
Permalink
I hope text file attachments are ok.

Logs are truncated to most recent entries.

Config file is complete.

Brian

Sent from my iPhone
Date: February 23, 2017 at 18:53:02 MST
Loading...