Updated: 29 September 2026 · Applies to: chrony 4.5 to 4.8 on AlmaLinux and Rocky Linux 9 and 10, RHEL 9 and 10, Ubuntu 24.04 and 26.04, Debian 13

chrony keeps the system clock in sync. For troubleshooting time problems, it can write detailed statistics to log files: every measurement from each time server, the statistics it calculates from them, and how it corrects the clock. We turned the logs on with chrony 4.8 on AlmaLinux 9 and read them below.

1. Enable the logs

The setting is already in the configuration file, commented out. Edit /etc/chrony.conf (AlmaLinux, Rocky Linux, RHEL) or /etc/chrony/chrony.conf (Ubuntu, Debian) and enable:

logdir /var/log/chrony
log measurements statistics tracking
  • measurements: each raw NTP measurement: server, stratum, offset, delay and the result of chrony's sanity tests.
  • statistics: what chrony calculated from the measurements of each server (estimated offset, frequency and how reliable they are).
  • tracking: the changes chrony made to the system clock and the current estimated error.

Other options are rtc (the hardware clock) and refclocks (local reference clocks such as GPS). To be warned about large clock jumps, add logchange 0.5: chrony then writes a syslog message when it corrects the clock by more than half a second.

2. Check and restart

chronyd -p | grep -E '^(log|logdir)'
systemctl restart chronyd    # "chrony" on Ubuntu and Debian

chronyd -p prints the configuration as chronyd reads it, without starting it.

3. Read the logs

ls /var/log/chrony/
tail -5 /var/log/chrony/measurements.log
tail -5 /var/log/chrony/tracking.log

In our test, measurements.log, statistics.log and tracking.log appeared within seconds. Each file starts with a header line that names the columns, for example Date (UTC) Time IP Address L St ... Offset Peer del. ... in the measurements log. Useful things to look for:

  • Large or jumping Offset values for one server only: that server is unreliable; remove it from the configuration.
  • A high Peer del. (delay): the network path to that server is slow or congested.
  • In tracking.log, a steadily growing Freq ppm: the clock drifts a lot, which happens on some virtual machines.

For a quick live view without logs, use chronyc tracking, chronyc sources -v and chronyc sourcestats.

Log rotation

The logs grow continuously. The chrony packages include a logrotate configuration for /var/log/chrony/*.log; check it in /etc/logrotate.d/chrony, and turn the statistics logs off again when you have finished troubleshooting.

Official documentation: chrony.conf manual.

Ucartz services for this topic

Was this answer helpful? 0 Users Found This Useful (0 Votes)