Environment
- Make target:
make build-node-ova-vsphere-photon-5
- Run using container image? (Y/N): N
- Environment vars:
Issue is reproducible with default settings
Issue is reproducible with default settings
What steps did you take and what happened?
-
Built a VM image.
-
Provisioned new Kubernetes nodes using the generated image.
-
When inspecting the NTP-services on a new node, I found that both chronyd and systemd-timesyncd were enabled. Chronyd is the active service initially. Systemd-timesyncd is inactive but remains enabled:
sudo systemctl status chronyd
● chronyd.service - NTP client/server
Loaded: loaded (/usr/lib/systemd/system/chronyd.service; enabled; preset: enabled)
Active: active (running) since Tue 2026-03-24 08:34:34 UTC; 2 weeks 0 days ago
Docs: man:chronyd(8)
man:chrony.conf(5)
Main PID: 448 (chronyd)
Tasks: 1 (limit: 23411)
Memory: 1.3M
CPU: 3.637s
CGroup: /system.slice/chronyd.service
└─448 /usr/sbin/chronyd
---
sudo systemctl status systemd-timesyncd
○ systemd-timesyncd.service - Network Time Synchronization
Loaded: loaded (/usr/lib/systemd/system/systemd-timesyncd.service; enabled; preset: enabled)
Active: inactive (dead)
Docs: man:systemd-timesyncd.service(8)
- On some nodes, what seems to be a network issue caused chrony to temporarily lose connection to its configured NTP-servers. As a result of this the chronyd service is stopped, and systemd-timesyncd is started in its place. This leads to an inconsistent state across the cluster where different nodes are using different NTP services.
sudo systemctl status chronyd
○ chronyd.service - NTP client/server
Loaded: loaded (/usr/lib/systemd/system/chronyd.service; enabled; preset: enabled)
Active: inactive (dead) since Tue 2026-03-31 08:19:39 UTC; 3min 12s ago
Duration: 1w 19h 41min 50.006s
Docs: man:chronyd(8)
man:chrony.conf(5)
Main PID: 424 (code=exited, status=0/SUCCESS)
CPU: 2.001s
chronyd[424]: Selected source 10.x.x.x
chronyd[424]: Can't synchronise: no selectable sources
chronyd[424]: Selected source 10.x.x.x
chronyd[424]: chronyd exiting
systemd[1]: Stopping NTP client/server...
systemd[1]: chronyd.service: Deactivated successfully.
systemd[1]: Stopped NTP client/server.
systemd[1]: chronyd.service: Consumed 2.001s CPU time.
---
sudo systemctl status systemd-timesyncd
● systemd-timesyncd.service - Network Time Synchronization
Loaded: loaded (/usr/lib/systemd/system/systemd-timesyncd.service; enabled; preset: enabled)
Active: active (running) since Tue 2026-03-31 08:19:39 UTC; 7min ago
Docs: man:systemd-timesyncd.service(8)
Main PID: 1042156 (systemd-timesyn)
Status: "Contacted time server 216.239.35.0:123 (time1.google.com)."
Tasks: 2 (limit: 19171)
Memory: 1.5M
CPU: 353ms
CGroup: /system.slice/systemd-timesyncd.service
└─1042156 /usr/lib/systemd/systemd-timesyncd
systemd[1]: Starting Network Time Synchronization...
systemd[1]: Started Network Time Synchronization.
systemd-timesyncd[1042156]: Contacted time server 216.239.35.0:123 (time1.google.com).
systemd-timesyncd[1042156]: Initial clock synchronization to Tue 2026-03-31 08:19:39.451626 UTC.
What did you expect to happen?
Chrony should be the active NTP-client while systemd-timesyncd should be disabled.
Relevant log output
Log Output
<!--
If you have any relevant build logs that could help debug this issue please include them here
but MAKE SURE ANY SENSITIVE INFO IS REMOVED!
-->
Anything else you would like to add?
/kind bug
Environment
make build-node-ova-vsphere-photon-5What steps did you take and what happened?
Built a VM image.
Provisioned new Kubernetes nodes using the generated image.
When inspecting the NTP-services on a new node, I found that both chronyd and systemd-timesyncd were enabled. Chronyd is the active service initially. Systemd-timesyncd is inactive but remains enabled:
What did you expect to happen?
Chrony should be the active NTP-client while systemd-timesyncd should be disabled.
Relevant log output
Log Output
Anything else you would like to add?
/kind bug