OPINIONATED OMARCHY
TROUBLESHOOTING CORPUS
790
RECORDS
790
AUDITED
0
UNCHECKED
2388
SOURCES
LIVE
apps-services medium occasional DISPUTE

Fix systemd-timesyncd never synchronising ("Timed out waiting for reply")

archomarchysystemdlaptop
CORRECTED An auditor found the fix wrong or incomplete and rewrote it. · checked against Omarchy omarchy 4.0.4-1 on 2026-10-05
AUDIT NOTE

Both cited forum threads were read. 270929 is solved as an IT policy blocking external NTP, which supports the first cause, and its log line with 146.164.48.5 (2.arch.pool.ntp.org) is the record's symptom. 274780 ends with a moderator saying ntpd and ntpdate were enabled alongside timesyncd and should be disabled, which supports the second cause as advice, but nothing in the thread shows a port conflict. timesyncd is an SNTP client that sends from an ephemeral port, so 'fight over the port' is an invented mechanism and is removed from the cause. The fix ran 'systemctl disable --now ntpd.service ntpdate.service chronyd.service' unconditionally, which names units that are usually not installed (on this 4.0.4-1 machine neither ntp nor chrony is installed) and makes systemctl report 'Unit file ... does not exist' and fail, so the corrected fix disables only the units the check showed as enabled. omarchy-update-time on 4.0.4-1 is exactly 'sudo systemctl restart systemd-timesyncd' and calls sudo itself, confirmed. Omarchy's ufw does not cause this: /etc/ufw/before.rules accepts RELATED,ESTABLISHED on input, so replies to outbound NTP pass, and that is added to the cause. The '<your-ntp-server>' placeholder was inside a copy-paste pipeline, so the corrected fix has the user edit the file instead. Not exercised: no service was changed.

The cause above was rewritten on 2026-10-05 to match this note; the fix was corrected by the audit itself.

Symptom

timedatectl status keeps saying System clock synchronized: no with NTP service: active, and the clock drifts. The timesyncd log repeats:


Timed out waiting for reply from 146.164.48.5:123 (2.arch.pool.ntp.org).

Cause

Two causes are reported. The network drops outbound UDP 123 to public servers (office or school networks, some ISPs), so the pool servers never answer. Or a second time daemon (ntpd, ntpdate or chronyd) is enabled alongside timesyncd, and an Arch forum moderator diagnosed exactly that combination in a thread with this symptom. Only one time service should manage the clock. Omarchy's own firewall is not the cause: ufw's default deny applies to new incoming connections, and /etc/ufw/before.rules accepts replies to traffic the machine sent.

Fix

Look at what timesyncd is doing:


timedatectl status
timedatectl timesync-status
journalctl -u systemd-timesyncd -b --no-pager | tail -20
systemctl is-enabled ntpd.service ntpdate.service chronyd.service 2>/dev/null

The last command prints a line only for units that are installed. If any of them prints enabled, keep one time service. To keep timesyncd, disable only the units that printed enabled. For example, if ntpd.service and ntpdate.service were enabled:


sudo systemctl disable --now ntpd.service ntpdate.service
sudo timedatectl set-ntp true

Do not list a unit that is not installed, because systemctl disable then fails with Unit file ... does not exist.

If the network blocks public NTP, ask whoever runs it which NTP server is allowed, then create a drop-in naming it:


sudo mkdir -p /etc/systemd/timesyncd.conf.d
sudoedit /etc/systemd/timesyncd.conf.d/local.conf

Put this in the file, with the allowed server's name or address after NTP=:


[Time]
NTP=ntp.your-network.example

Restart timesyncd. On Omarchy 4 omarchy-update-time does exactly this restart and calls sudo itself:


omarchy-update-time                              # Omarchy 4
sudo systemctl restart systemd-timesyncd        # plain Arch

Verify

timedatectl timesync-status shows a server and a nonzero Packet count, and timedatectl status reports System clock synchronized: yes.

Sources

timesyncd-timed-out-waiting-for-reply-ntp-blocked