Re: [chrony-users] Questions about "holdover" state

[Miroslav Lichvar <mlichvar@xxxxxxxxxx>]

On Mon, Jan 12, 2026 at 11:29:24AM +0000, Valera Requena, Juan wrote:
>   1.
> Identification of holdover state
> Is there a way to explicitly identify that chronyd is operating in a holdover state using chronyc or chronyd status information?
> For example:
>      *   Is there a specific flag, status field, or command output that indicates “synchronized but no reachable sources”?

No.

>      *   Should this be inferred indirectly (e.g., from source reachability, Leap status, or Reference ID)?

Yes, the sources command prints the reachability of individual
sources. If the selected sources have 0 reachability, it means the
clock was not updated in last 8 polls. You could also check the
timestamp of the last reference update from the tracking report to see
if it's not too old depending on your requirements.

>   2.
> Transition from holdover to unsynchronized
> How does chronyd internally determine when a system, while in holdover, transitions to a non-synchronized state?
>      *   Is this based on a maximum estimated error, time since last valid measurement, root dispersion growth, or another threshold?
>      *   Are there configuration parameters that influence this decision (e.g., limits on acceptable uncertainty or timeout values)?

It doesn't switch to an unsynchronized state. It's up to the clients
to decide what is still acceptable for them by inspecting the time of
the last reference update, an the growing root delay and dispersion.

-- 
Miroslav Lichvar


-- 
To unsubscribe email chrony-users-request@xxxxxxxxxxxxxxxxxxxx 
with "unsubscribe" in the subject.
For help email chrony-users-request@xxxxxxxxxxxxxxxxxxxx 
with "help" in the subject.
Trouble?  Email listmaster@xxxxxxxxxxxxxxxxxxxx.