Chrony¶
This page is the starting point for chrony, the chrony role: one container, chrony, on the compute, control,
monitoring, network and storage nodes of the example inventory, selected by the tag chrony. Its play is the first play
for every host (Tags §3, row chrony), and -t chrony is
the recommended way to push common changes (Apply common changes (fluentd, cron, kolla_toolbox)
§3). The lists below say where the study
covers how to operate it, its impact, its dependencies and the evidence, what to check around a run and what stays open;
the service card follows them.
Where to look¶
Operate¶
- Scenario: Apply
commonchanges (fluentd, cron, kolla_toolbox) §3 (the way to pushcommonchanges) - Command: Apply
commonchanges (fluentd, cron, kolla_toolbox) §1.1 - Command quick lookup by tag, row
-t chrony reconfigure: hosts touched with the default inventory: all hosts; can restart, only if drift is detected: chrony + common (fluentd, cron, toolbox); typical duration driver: the number of hosts.
Impact and risk¶
Dependencies¶
- Dependencies: Component dependencies §2, Component dependencies §5
Evidence¶
- Fact sheet: fact sheet A §4
- Image side: Kolla images §5 (row
chrony)
Before and after a run¶
- Tags §3.1, row
chrony: safe on its own "yes; also the cheapest way to push common config fleet-wide"; prerequisite: VIP up for the chrony clients. - Tags §3, row
chrony: the containers that can restart, only if a change is detected, are chrony, fluentd, cron and kolla_toolbox. - Apply
commonchanges (fluentd, cron, kolla_toolbox) §1.1:-t chrony reconfigurealso reconfigures chrony on those hosts. Ifenable_chronyisno, use any service tag whose play covers the hosts, e.g.-t openvswitchfor network and compute hosts. - Apply
commonchanges (fluentd, cron, kolla_toolbox) §4: the verification before and after, forfluentd,cronandkolla_toolbox.
Known limits and open points¶
- fact sheet A §4.8, "Operationally notable", four items: low data-plane risk (NOT AVAILABLE: whether a restart steps the clock); cost scales with the number of hosts; chrony.conf depends on the haproxy group; the VIP is the internal time source (NOT AVAILABLE: whether reachable external NTP servers mitigate the change of clock source when the VIP fails over).
- fact sheet A §4.7: NOT AVAILABLE: whether, on a first
deploy, chronyd on a non-VIP chrony-server host can bind the VIP before
ip_nonlocal_bindis set, and whether it then serves before a later restart. - fact sheet D §5.12: by default every compute queries public pool servers. NOT AVAILABLE: whether a chrony restart (image or config change) on a drifted compute steps the clock by more than 3 s in the first updates.
- R14: a
bootstrap-serversre-run on a live host removes host packages (open-iscsi,libvirt*,chrony, …). - What is not known: the fact sheets mark their own NOT AVAILABLE points inline; the table lists fact sheet A §4.7 and §4.8 and fact sheet D §5.12 among the sections that contain such marks.
- Known bugs and limitations, Open questions and Findings have no entry on chrony.
Service card¶
The rows of the card are explained in How to read a service card.
| Item | Value |
|---|---|
| Play | site.yml:114 |
| Tags | chrony |
| Play host patterns | chrony-server, chrony |
| Node types in play (example inventory) | compute, control, monitoring, network, storage |
| Node types running its containers | compute, control, monitoring, network, storage |
| Containers | chrony |
deploy/reconfigure flow (tasks/) |
config → check-containers → flush handlers |
| Task files | check-containers.yml, check.yml, config.yml, deploy-containers.yml, deploy.yml, main.yml, precheck.yml, pull.yml, reconfigure.yml, stop.yml, upgrade.yml |
| Templates | chrony.conf.j2, chrony.json.j2 |
| Handlers (definition order) | Restart chrony container |
| Key variables | enable_chrony, external_ntp_servers |
| Role dependencies | common (meta) |
Related¶
- Apply
commonchanges (fluentd, cron, kolla_toolbox) —-t chrony reconfigure, the command that pushescommonchanges - Common: fluentd, cron, kolla_toolbox — the containers that a
-t chronyrun also checks - HAProxy + Keepalived — in the example inventory
chrony-serveris the haproxy group, and the VIP is the internal time source - Component dependencies §5 — the global inputs of a compute's
chrony.conf - Components — the other component pages