| RE: [chrony-users] chrony switch source issue |
[ Thread Index |
Date Index
| More chrony.tuxfamily.org/chrony-users Archives
]
- To: "chrony-users@xxxxxxxxxxxxxxxxxxxx" <chrony-users@xxxxxxxxxxxxxxxxxxxx>
- Subject: RE: [chrony-users] chrony switch source issue
- From: "Bernstein, Amit" <amitbern@xxxxxxxxxx>
- Date: Tue, 3 Jan 2023 07:41:21 +0000
- Accept-language: en-US
- Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=amazon.com; i=@amazon.com; q=dns/txt; s=amazon201209; t=1672731687; x=1704267687; h=from:to:date:message-id:references: content-transfer-encoding:mime-version:subject; bh=i6A7vabpHVkapq0kVTh2PoGnI+6HSPIqDPWhSLFC77k=; b=vjkFaeK6ObMTE6IC8hZUMcTIx2zt1yIwFK3yC7k8weybJXWih49rHtyp eyenIaUuOMNq6KRPuHdBXNzVVQEyzexeYxfPTKXrbEGoixsBiCwOqg0ZI NcLqc9kzRir7Jg5NOwS/FwO1GxTHM68nMQotboJNrkQaMlhEsm8ldXCyq 0=;
- Thread-index: AQHZHrisyF47EWUvE0aGYybhJhG8aK6LOIoAgADyodCAACR6EA==
- Thread-topic: [chrony-users] chrony switch source issue
And BTW, its still running for 18 hours and still chrony didn't change sources
MS Name/IP address Stratum Poll Reach LastRx Last sample
===============================================================================
#? PHC0 0 4 0 18h -1284us[ -150ns] +/- 87ns
^+ gotoro.hojmark.net 2 10 377 784 -1645us[-1645us] +/- 61ms
^+ animine.org 2 10 377 446 -1107us[-1107us] +/- 72ms
^+ mx.ack512.net 2 10 377 814 +4038us[+4038us] +/- 104ms
^* t2.time.bf1.yahoo.com 2 10 377 815 +2050us[+2079us] +/- 38ms
Name/IP Address NP NR Span Frequency Freq Skew Offset Std Dev
==============================================================================
PHC0 7 3 86 +0.010 0.010 -621us 91ns
gotoro.hojmark.net 64 34 17h +0.005 0.015 -1607us 663us
animine.org 33 19 552m +0.005 0.043 -1994us 676us
mx.ack512.net 64 33 17h -0.001 0.009 +4347us 381us
t2.time.bf1.yahoo.com 64 35 17h +0.001 0.015 +1186us 632us
Remote address : 147.182.158.78 (93B69E4E)
Remote port : 123
Local address : 172.31.42.144 (AC1F2A90)
Leap status : Normal
Version : 4
Mode : Server
Stratum : 2
Poll interval : 10 (1024 seconds)
Precision : -22 (0.000000238 seconds)
Root delay : 0.001709 seconds
Root dispersion : 0.029678 seconds
Reference ID : 61B7CE58 ()
Reference time : Tue Jan 03 07:10:54 2023
Offset : +0.001645330 seconds
Peer delay : 0.060395706 seconds
Peer dispersion : 0.000000261 seconds
Response time : 0.000067787 seconds
Jitter asymmetry: +0.00
NTP tests : 111 111 1111
Interleaved : No
Authenticated : No
TX timestamping : Kernel
RX timestamping : Kernel
Total TX : 89
Total RX : 89
Total valid RX : 89
Remote address : 45.33.65.68 (2D214144)
Remote port : 123
Local address : 172.31.42.144 (AC1F2A90)
Leap status : Normal
Version : 4
Mode : Server
Stratum : 2
Poll interval : 10 (1024 seconds)
Precision : -24 (0.000000060 seconds)
Root delay : 0.023773 seconds
Root dispersion : 0.026886 seconds
Reference ID : D9B4D1D6 ()
Reference time : Tue Jan 03 07:24:09 2023
Offset : +0.001106975 seconds
Peer delay : 0.065741964 seconds
Peer dispersion : 0.000000085 seconds
Response time : 0.000035461 seconds
Jitter asymmetry: +0.43
NTP tests : 111 111 1111
Interleaved : No
Authenticated : No
TX timestamping : Kernel
RX timestamping : Kernel
Total TX : 117
Total RX : 117
Total valid RX : 117
Remote address : 136.243.229.3 (88F3E503)
Remote port : 123
Local address : 172.31.42.144 (AC1F2A90)
Leap status : Normal
Version : 4
Mode : Server
Stratum : 2
Poll interval : 10 (1024 seconds)
Precision : -24 (0.000000060 seconds)
Root delay : 0.011658 seconds
Root dispersion : 0.009628 seconds
Reference ID : ED11CC5F ()
Reference time : Tue Jan 03 07:18:39 2023
Offset : -0.004037738 seconds
Peer delay : 0.176519439 seconds
Peer dispersion : 0.000000083 seconds
Response time : 0.000013426 seconds
Jitter asymmetry: +0.00
NTP tests : 111 111 1111
Interleaved : No
Authenticated : No
TX timestamping : Kernel
RX timestamping : Kernel
Total TX : 89
Total RX : 89
Total valid RX : 89
Remote address : 72.30.35.88 (481E2358)
Remote port : 123
Local address : 172.31.42.144 (AC1F2A90)
Leap status : Normal
Version : 4
Mode : Server
Stratum : 2
Poll interval : 10 (1024 seconds)
Precision : -25 (0.000000030 seconds)
Root delay : 0.000107 seconds
Root dispersion : 0.000397 seconds
Reference ID : 628B853E ()
Reference time : Tue Jan 03 07:20:10 2023
Offset : -0.002078954 seconds
Peer delay : 0.075100102 seconds
Peer dispersion : 0.000000053 seconds
Response time : 0.000008782 seconds
Jitter asymmetry: +0.00
NTP tests : 111 111 1111
Interleaved : No
Authenticated : No
TX timestamping : Kernel
RX timestamping : Kernel
Total TX : 89
Total RX : 89
Total valid RX : 89
S Name/IP Address Auth COpts EOpts Last Score Interval Leap
=======================================================================
S PHC0 N ----- ----- 17h 1.0 -66ms +64ms N
+ gotoro.hojmark.net N ----- ----- 337 1.0 -49ms +45ms N
+ animine.org N ----- ----- 0 1.0 -47ms +53ms N
+ mx.ack512.net N ----- ----- 368 1.0 -94ms +104ms N
* t2.time.bf1.yahoo.com N ----- ----- 369 1.0 -36ms +38ms N
-----Original Message-----
From: Bernstein, Amit
Sent: Tuesday, January 3, 2023 07:38
To: chrony-users@xxxxxxxxxxxxxxxxxxxx
Subject: RE: [EXTERNAL][chrony-users] chrony switch source issue
I understand, but the behavior doesn't seem right, PHC device with nsec accuracy returns an error for 1 hour, and chrony refuses to replace the source to a usec accuracy NTP source?
This means chrony still "thinks" that an error response to a PHC device retrieved for 1 hour has better accuracy than NTP? That doesn't make sense.
If I set the PHC device to return zero instead of a real time, it takes couple of seconds for chrony to identify the timing issue and replace the source to NTP. I think this behavior should occur with PHC device error also.
Maybe the "smallest interval" you mention is wrongly calculated on PHC error?
Or maybe the system clock which was synced to the PHC at first wasn't changed too much (small shift than NTP) after 1 hour of not getting PHC timestamps?
Thanks for assisting.
-----Original Message-----
From: Miroslav Lichvar <mlichvar@xxxxxxxxxx>
Sent: Monday, January 2, 2023 17:01
To: chrony-users@xxxxxxxxxxxxxxxxxxxx
Subject: RE: [EXTERNAL][chrony-users] chrony switch source issue
CAUTION: This email originated from outside of the organization. Do not click links or open attachments unless you can confirm the sender and know the content is safe.
On Mon, Jan 02, 2023 at 02:42:59PM +0000, Bernstein, Amit wrote:
> Reproduced the issue for 60 minutes
>
> $chronyc sources
> MS Name/IP address Stratum Poll Reach LastRx Last sample
> ===============================================================================
> #* PHC0 0 4 0 60m -126ns[ -150ns] +/- 87ns
> ^- gotoro.hojmark.net 2 9 377 38 +201us[ +201us] +/- 67ms
> ^- animine.org 2 8 375 242 +244us[ +244us] +/- 78ms
> ^- mx.ack512.net 2 9 377 45 +5665us[+5665us] +/- 104ms
> ^- t2.time.bf1.yahoo.com 2 9 377 35 +2039us[+2039us] +/- 38ms
The estimated maximum error of the PHC0 source here is best by 6 orders of magnitude. If it becomes unreachable, you woudln't normally want to immediately switch to the next best source. It's nanoseconds vs milliseconds.
> S Name/IP Address Auth COpts EOpts Last Score Interval Leap
> =======================================================================
> * PHC0 N ----- ----- 63m 1.0 -3745us +3744us N
> D gotoro.hojmark.net N ----- ----- 185 1.0 -53ms +52ms N
> D animine.org N ----- ----- 132 1.0 -69ms +73ms N
> D mx.ack512.net N ----- ----- 191 1.0 -99ms +110ms N
> D t2.time.bf1.yahoo.com N ----- ----- 182 1.0 -35ms +40ms N
After 1 hour it still has the smallest interval, so it is working as intended.
The FAQ has some suggestions on how to force a faster reselection if you need that.
--
Miroslav Lichvar
--
To unsubscribe email chrony-users-request@xxxxxxxxxxxxxxxxxxxx
with "unsubscribe" in the subject.
For help email chrony-users-request@xxxxxxxxxxxxxxxxxxxx
with "help" in the subject.
Trouble? Email listmaster@xxxxxxxxxxxxxxxxxxxx.
--
To unsubscribe email chrony-users-request@xxxxxxxxxxxxxxxxxxxx
with "unsubscribe" in the subject.
For help email chrony-users-request@xxxxxxxxxxxxxxxxxxxx
with "help" in the subject.
Trouble? Email listmaster@xxxxxxxxxxxxxxxxxxxx.