ChronyD

Updated 29 October 2025

Introduction

Commands Details

Clock Offset “chronyc tracking”

We can use the following command to check for the Clock Offset:

bash

chronyc tracking

Its output is:

bash

[root@GW01A monitoring]# chronyc tracking
Reference ID    : 0A0E9D45 (10.14.157.69)
Stratum         : 2
Ref time (UTC)  : Wed Oct 29 07:55:35 2025
System time     : 0.000391756 seconds fast of NTP time
Last offset     : +0.000847606 seconds
RMS offset      : 0.000840737 seconds
Frequency       : 111.489 ppm slow
Residual freq   : +0.107 ppm
Skew            : 0.615 ppm
Root delay      : 0.002051588 seconds
Root dispersion : 0.003422280 seconds
Update interval : 1801.7 seconds
Leap status     : Normal

This command has the fields with its explanation:

Fields Explanation

Field	Example	Meaning	Good Range / When to Alert
Reference ID	0A0E9D45 (10.14.157.69)	Which server the system is synced to	Should not be 127.127.x.x (local clock)
Stratum	2	How many layers away from real time (UTC)	✅ 1–4 good⚠️ 5–10 weak🔔 >10 bad
Ref time (UTC)	Wed Oct 29 11:20:45 2025	Time of last sync update	Should be recentAlert if too old
System time	0.000142867 seconds fast of NTP time	Actual difference between system clock & NTP clock	✅ < 0.5 sec🔔 > 1 sec
Last offset	+0.000144842 seconds	Most recent measured time drift	✅ < 0.1 sec🔔 > 0.5 sec
RMS offset	0.000229859 seconds	Long-term average drift	Lower = stableAlert if > 0.5 sec
Frequency	114.245 ppm slow	How much system clock is adjusting rate	High value may indicate unstable hardware
Residual freq	+0.003 ppm	Remaining frequency adjustment	Typically small → OK
Skew	0.079 ppm	Expected error in frequency	Low = stable
Root delay	0.000943802 seconds	Total network delay to server	✅ lower is betterAlert if > 0.1 sec
Root dispersion	0.001699327 seconds	Drift error accumulated from network + server	✅ < 1 secAlert if > 3 sec
Update interval	3094.3 seconds	Time between sync polls	Not an alert field unless too large (> 1 hr)
Leap status	Normal	Leap second insertion status	✅ Must be Normal🔔 Not synchronized

Sync status summary “chronyc sources”

We can use the following command to check for the clock sync status:

bash

chronyc sources

The output is:

bash

chronyc sources
MS Name/IP address         Stratum Poll Reach LastRx Last sample               
===============================================================================
^* 10.14.157.69                  1  10   375   871  +2871us[+3016us] +/- 2859us

Fields Explanation

The fields in the output of this command can be explained as follows:

Column	Example	Meaning	Good Value / Alert Condition
M (Mode)	^	Connection type:^ server, = peer, # local clock	Not an alert field
S (State)	*	Sync health:* = active source+ = backup- = not combined? = unreachablex error~ variable	✅ Must have at least one 🔔 Alert if no present
Name/IP Address	10.14.157.69	The NTP server used	Not an alert field
Stratum	1	Distance from UTC sourceLower is more accurate	✅ 1–4 good⚠️ 5–15 weak🔔 Alert if > 10
Poll	10	Poll interval = 2^Poll seconds10 → 1024s (~17 min)	Large values mean slower updatesNo direct alert
Reach	375	Octal history of last 8 polls (1=success)	✅ ≥ 1 means reachable🔔 Alert if 0 (unreachable)
LastRx	923	Seconds since last successful response	✅ Under 1000 typically good🔔 Alert if too old
Last sample – adjusted offset	+2871us	How much system time adjusted	✅ < 100 ms⚠️ 100–500 ms🔔 Critical > 500 ms
Last sample – measured offset	[+3016us]	Offset measured from server	Similar thresholds as adjusted offset
Estimated error (jitter)	+/- 2859us	Stability of measurement	Lower = better🔔 Alert if > 10 ms

Spotted a mistake or want something added? Send me a note.

📄ChronyD

ChronyD