EN FR
Check-NTP / Daemon Troubleshooting

NTP Daemon Troubleshooting: Fix ntpd, chronyd & systemd-timesyncd

Troubleshoot ntpd, chronyd, and systemd-timesyncd problems

Quick Navigation

Identify Your NTP Daemon

First, determine which NTP daemon your system uses:

# Check which NTP service is active root@server:~# systemctl list-units --type=service | grep -E "ntp|chrony|time" chronyd.service loaded active running NTP client/server systemd-timesyncd.service loaded active running Network Time Synchronization
DaemonConfig FileQuery Command
ntpd (classic) /etc/ntp.conf ntpq -p
chronyd /etc/chrony/chrony.conf or /etc/chrony.conf chronyc sources
systemd-timesyncd /etc/systemd/timesyncd.conf timedatectl timesync-status
W32Time (Windows) Registry / GPO w32tm /query /status

Daemon Not Starting

Symptom: Service fails to start or immediately stops

The NTP service won't start, or starts and immediately crashes.

Check Service Status

root@server:~# systemctl status chronyd ● chronyd.service - NTP client/server Loaded: loaded (/usr/lib/systemd/system/chronyd.service; enabled) Active: failed (Result: exit-code) since... # Check detailed logs root@server:~# journalctl -xeu chronyd --no-pager | tail -50

Common Causes & Solutions

Cause 1: Configuration Syntax Error

# Test config file syntax (chronyd) root@server:~# chronyd -p # Test config file syntax (ntpd) root@server:~# ntpd -n -d -c /etc/ntp.conf

Cause 2: Port 123 Already in Use

# Check what's using port 123 root@server:~# ss -ulnp | grep 123 UNCONN 0 0 *:123 *:* users:(("ntpd",pid=1234,fd=16)) # Kill conflicting process root@server:~# systemctl stop ntpd root@server:~# systemctl start chronyd

Cause 3: Permission Issues

# Check drift file permissions root@server:~# ls -la /var/lib/chrony/drift root@server:~# chown chrony:chrony /var/lib/chrony/drift # For ntpd root@server:~# chown ntp:ntp /var/lib/ntp/drift

Cause 4: Multiple NTP Services Conflicting

# Disable conflicting services root@server:~# systemctl disable --now systemd-timesyncd root@server:~# systemctl disable --now ntpd # Enable only one root@server:~# systemctl enable --now chronyd

Stratum 16 / No Sync Source

Symptom: System shows stratum 16 or "unsynchronised"

Stratum 16 means the NTP daemon has no valid time source. This is the "invalid/unknown" stratum level.

# ntpq showing stratum 16 root@server:~# ntpq -c rv ...stratum=16, precision=-23... # chronyc showing not synchronized root@server:~# chronyc tracking Reference ID : 00000000 () Stratum : 0 ... Leap status : Not synchronised

Solutions

Check NTP Server Configuration

# Verify servers are configured root@server:~# grep -E "^server|^pool" /etc/chrony/chrony.conf pool 2.debian.pool.ntp.org iburst server ntp.rdem-systems.com iburst prefer

Test Server Connectivity

# Can we reach NTP servers? root@server:~# chronyc sources -v MS Name/IP address Stratum Poll Reach LastRx Last sample =============================================================================== ^? ntp.rdem-systems.com 0 6 0 - +0ns[ +0ns] +/- 0ns # Reach = 0 means no successful queries

Check Firewall

# Test UDP 123 connectivity root@server:~# nc -vzu ntp.rdem-systems.com 123 # If blocked, see firewall guide

See Firewall Configuration Guide →

Force Initial Sync

# For chronyd - force step sync root@server:~# chronyc makestep # For ntpd - manual initial sync root@server:~# systemctl stop ntpd root@server:~# ntpdate -b pool.ntp.org root@server:~# systemctl start ntpd

Stuck in .INIT State

Symptom: NTP shows .INIT for all servers

The refid column shows ".INIT." indicating the daemon hasn't completed initial synchronization.

root@server:~# ntpq -p remote refid st t when poll reach delay offset jitter ============================================================================== 0.pool.ntp.org .INIT. 16 p - 64 0 0.000 0.000 0.000

Common Causes

Solution: Check DNS Resolution

# Test DNS for NTP pool root@server:~# host pool.ntp.org pool.ntp.org has address 162.159.200.1 pool.ntp.org has address 162.159.200.123 # If DNS fails, use IP addresses in config root@server:~# echo "server 195.154.XXX.XXX iburst" >> /etc/ntp.conf

Solution: Fix Large Time Offset

# Check current offset root@server:~# ntpdate -q pool.ntp.org server 162.159.200.1, stratum 3, offset -3600.123456... # If offset > 1000s, manually set time first root@server:~# systemctl stop ntpd root@server:~# ntpdate -b pool.ntp.org root@server:~# systemctl start ntpd

Wait Time

Initial synchronization can take 5-15 minutes depending on network conditions and polling intervals. Check the "reach" value - it should increase from 0 to 377 over time as queries succeed.

False Ticker Detection

Symptom: Server marked with 'x' (falseticker)

NTP has determined this server's time differs significantly from the majority of sources.

root@server:~# ntpq -p remote refid st t when poll reach delay offset jitter ============================================================================== xntp1.example.com .GPS. 1 u 45 64 377 12.543 -250.284 1.023 *ntp2.example.com .PPS. 1 u 42 64 377 28.123 0.842 2.156

Understanding False Tickers

NTP uses a voting algorithm. If one server disagrees with the majority by more than a threshold, it's marked as a "false ticker" (x) and excluded from time calculations.

Solution 1: Add More Servers

With only 2 servers, NTP can't determine which is correct. Use at least 4 servers:

# /etc/ntp.conf or /etc/chrony/chrony.conf server ntp.rdem-systems.com iburst prefer pool 0.pool.ntp.org iburst maxsources 2 pool 1.pool.ntp.org iburst maxsources 2

Solution 2: Verify Server Accuracy

# Query each server individually root@server:~# ntpdate -q ntp1.example.com root@server:~# ntpdate -q ntp2.example.com # Remove servers with consistently bad offsets

Solution 3: Check GPS/PPS Source

If a GPS receiver is marked false ticker, check:

  • GPS antenna signal strength
  • Number of satellites locked (need 4+)
  • PPS signal cable and configuration

Reach Value is 0

Symptom: Reach column shows 0 for all servers

No successful NTP queries have been received. The daemon cannot communicate with any server.

root@server:~# ntpq -p remote refid st t when poll reach delay offset jitter ============================================================================== pool.ntp.org .POOL. 16 p - 64 0 0.000 0.000 0.000

Understanding Reach

Reach is an 8-bit octal register showing the last 8 query results. 377 (octal) = 11111111 (binary) = all 8 queries succeeded. 0 = no successful queries.

Diagnostic Steps

# 1. Check daemon is sending queries root@server:~# tcpdump -i any port 123 -c 10 # 2. Check firewall isn't blocking root@server:~# iptables -L -n | grep 123 # 3. Test manual NTP query root@server:~# ntpdate -d pool.ntp.org 2>&1 | head -30 # 4. Check for routing issues root@server:~# traceroute -U -p 123 pool.ntp.org

Common Fixes

  1. Open firewall: ufw allow 123/udp
  2. Check DNS works: host pool.ntp.org
  3. Use IP instead of hostname in config
  4. Restart daemon: systemctl restart chronyd
  5. Wait - reach increases every poll interval

Chrony-Specific Issues

chronyc "506 Cannot talk to daemon"

Symptom: chronyc commands fail with connection error

Solution

# Check if chronyd is running root@server:~# systemctl status chronyd # Check cmdport setting in config root@server:~# grep cmdport /etc/chrony/chrony.conf # Ensure cmdport is not set to 0 (disabled)

"makestep" Doesn't Work

Check makestep configuration

# In /etc/chrony/chrony.conf, ensure makestep is configured makestep 1 3 # Step if offset > 1s, first 3 updates # Or force with -1 makestep 1 -1 # Step if offset > 1s, unlimited times

Slow Initial Sync

Speed up initial synchronization

# Add iburst to server lines server ntp.rdem-systems.com iburst # Use initstepslew for boot-time correction initstepslew 10 ntp.rdem-systems.com pool.ntp.org

Windows W32Time Issues

"The computer did not resync because no time data was available"

Solution

:: Reset W32Time configuration C:\> net stop w32time C:\> w32tm /unregister C:\> w32tm /register C:\> net start w32time :: Configure NTP server C:\> w32tm /config /manualpeerlist:"pool-ntp.rdem-systems.com,0x8" /syncfromflags:manual /update :: Force sync C:\> w32tm /resync /force

Domain Controller Time Issues

Configure PDC Emulator

:: On PDC Emulator (run as Administrator) C:\> w32tm /config /manualpeerlist:"pool-ntp.rdem-systems.com" /syncfromflags:manual /reliable:yes /update C:\> net stop w32time && net start w32time C:\> w32tm /resync :: Other DCs will sync from PDC automatically

Hyper-V Time Sync Conflicts

Virtual Machine Warning

Hyper-V VMs have time sync integration services that can conflict with W32Time. Disable VM time sync if using external NTP:

:: In Hyper-V Manager: :: VM Settings → Integration Services → Uncheck "Time synchronization" :: Or via PowerShell PS C:\> Disable-VMIntegrationService -VMName "YourVM" -Name "Time Synchronization"

KVM / Proxmox VE Time Sync

Proxmox VE Virtual Machines

KVM guests on Proxmox VE can experience significant clock drift if NTP is not properly configured inside the VM. The paravirtualized clock (kvm-clock) helps, but a dedicated NTP client remains essential for production workloads.

# Check if kvm-clock is active (inside the VM) root@vm:~# cat /sys/devices/system/clocksource/clocksource0/current_clocksource kvm-clock # Configure chrony for reliable sync inside a Proxmox VM root@vm:~# cat /etc/chrony/chrony.conf server ntp1.rdem-systems.fr iburst server ntp2.rdem-systems.fr iburst makestep 1 3 rtcsync # Restart chrony root@vm:~# systemctl restart chronyd && chronyc tracking

Complete guide: Hosting a public NTP server on a Proxmox VM →

Verify Your Fix

After troubleshooting, test your NTP synchronization:

Run NTP Diagnostics Firewall Guide

See also: NTS compatibility test · NTP server validator

check-ntp.net · NTP diagnostic by RDEM Systems

Home | Firewall Guide | Daemon Troubleshooting

Legal