First off I've searched for the answer here and it's not the standard disable/purge ntp/chrony. I can run dpkg -l | grep ntp and dpkg -l | grep chrony and neither returns anything, so I know it's not a conflict with that. It's also a fresh vagrant image of ubuntu 20.04 that has no cruft leftover from ugprading from a previous version of ubuntu.
With systemd-timesyncd enabled, at bootup it fails to start and sync time.
systemctl status systemd-timesyncd
● systemd-timesyncd.service - Network Time Synchronization
Loaded: loaded (/lib/systemd/system/systemd-timesyncd.service; enabled; vendor preset: enabled)
Active: inactive (dead)
Docs: man:systemd-timesyncd.service(8)
There are no conditionals that should block it in /lib/systemd/system/systemd-timesyncd.service, and it's all defaults
[Unit]
Description=Network Time Synchronization
Documentation=man:systemd-timesyncd.service(8)
ConditionCapability=CAP_SYS_TIME
ConditionVirtualization=!container
DefaultDependencies=no
After=systemd-sysusers.service
Before=time-set.target sysinit.target shutdown.target
Conflicts=shutdown.target
Wants=time-set.target time-sync.target
[Service]
AmbientCapabilities=CAP_SYS_TIME
CapabilityBoundingSet=CAP_SYS_TIME
ExecStart=!!/lib/systemd/systemd-timesyncd
LockPersonality=yes
MemoryDenyWriteExecute=yes
NoNewPrivileges=yes
PrivateDevices=yes
PrivateTmp=yes
ProtectControlGroups=yes
ProtectHome=yes
ProtectHostname=yes
ProtectKernelModules=yes
ProtectKernelTunables=yes
ProtectKernelLogs=yes
ProtectSystem=strict
Restart=always
RestartSec=0
RestrictAddressFamilies=AF_UNIX AF_INET AF_INET6
RestrictNamespaces=yes
RestrictRealtime=yes
RestrictSUIDSGID=yes
RuntimeDirectory=systemd/timesync
StateDirectory=systemd/timesync
SystemCallArchitectures=native
SystemCallErrorNumber=EPERM
SystemCallFilter=@system-service @clock
Type=notify
User=systemd-timesync
WatchdogSec=3min
[Install]
WantedBy=sysinit.target
Alias=dbus-org.freedesktop.timesync1.service
/etc/systemd/timesyncd.conf only has the following
[Time]
NTP=0.us.pool.ntp.org
FallbackNTP=ntp.ubuntu.com 1.us.pool.ntp.org time.cloudflare.com
But when I manually start it with systemctl restart systemd-timesyncd it works just fine, it only gets into the dead state at bootup.
systemctl restart systemd-timesyncd
systemctl status systemd-timesyncd
● systemd-timesyncd.service - Network Time Synchronization
Loaded: loaded (/lib/systemd/system/systemd-timesyncd.service; enabled; vendor preset: enabled)
Active: active (running) since Thu 2021-07-29 20:34:51 UTC; 6s ago
Docs: man:systemd-timesyncd.service(8)
Main PID: 1347 (systemd-timesyn)
Status: "Initial synchronization to time server 23.175.208.10:123 (0.us.pool.ntp.org)."
Tasks: 2 (limit: 2320)
Memory: 1.3M
CGroup: /system.slice/systemd-timesyncd.service
└─1347 /lib/systemd/systemd-timesyncd
I can even put systemctl restart systemd-timesyncd into rc.local and it works on bootup after rc.local runs. I could go that route and call it good, but it definitely should be working without that. I can't find any indication of failures in /var/log/syslog, either.
If it matters, this is a virtualbox vm, but I need to set this up on ec2 eventually.
