With systemd-243~rc1 in rawhide, which supports the original ntp-units.d/ syntax, the raison d'être for timedatex is gone. The is no problem with it being present, but I think it'd be nicer to have one package less. If you are OK with that, I'd add "Obsoletes:timedatex<0.7, Provides:timedatex=0.6" in the next build, and than we timedatex can be retired. timedatex is not Required by any package afaics, but comps requires timedatex, so the Provides in systemd would be necessary. I'll submit a PR to remove timedatex from comps too.
The chrony and ntp packages have "Recommends: timedatex". Those should probably be removed. I tried systemd-timedated from the latest systemd package and with all three NTP providers installed. There seem to be few issues. It would be nice to fix them before retiring timedatex to avoid regressions. "timedatectl set-ntp true" actives a wrong service (by priority) if it is already enabled. It doesn't enable the correct service and it doesn't disable the rest. timedatectl shows "System clock synchronized: no" when chronyd is running without "rtcsync". IIRC this is just a different interpretation of the data returned by adjtimex(). I'll submit a patch.
PR for the issue with reported synchronization status: https://github.com/systemd/systemd/pull/13245
> The chrony and ntp packages have "Recommends: timedatex". Those should probably be removed. They'll be harmless if timedatex is not available, so we can just remove them whenever. I'll file PRs when the other stuff is done. > timedatectl shows "System clock synchronized: no" when chronyd is running without "rtcsync". IIRC this is just a different interpretation of the data returned by adjtimex(). I'll submit a patch. Thanks. It is merged and in systemd-243-rc1. > "timedatectl set-ntp true" actives a wrong service (by priority) if it is already enabled. It doesn't enable the correct service and it doesn't disable the rest. That seems to be by design. I think that the idea is that if a service was already enabled previously, then we want to keep stability, and not start a different unit just because a different package was installed. If the user wants to do that, they should do 'set-ntp off, set-ntp on'. Is there some strong reason to prefer different behaviour? (I wrote some patches to add more verbose logging to timesyncd, so that it's easier to see what is going on and why. I'll submit that as PR later.)
(In reply to Zbigniew Jędrzejewski-Szmek from comment #3) > That seems to be by design. I think that the idea is that if a service was > already enabled previously, > then we want to keep stability, and not start a different unit just because > a different package was > installed. If the user wants to do that, they should do 'set-ntp off, > set-ntp on'. > Is there some strong reason to prefer different behaviour? Yes, I think that's confusing behavior. Setting ntp off and on should give the same result as just setting it on. There should be no stickiness (hidden state). It should behave like running an ansible role. The other service might have been enabled because no other was installed at the time. If I install a new NTP service with a higher priority, setting ntp on should enable and start the new service, and disable all other. Also, if for some reason multiple services are enabled, after setting ntp on there should be only one enabled to avoid systemd selecting between them randomly on the next boot.
https://github.com/systemd/systemd/pull/13265
Thanks for your work on this.
This bug appears to have been reported against 'rawhide' during the Fedora 31 development cycle. Changing version to '31'.
This bug appears to have been reported against 'rawhide' during the Fedora 31 development cycle. Changing version to 31.
FEDORA-2019-3d6348341f has been submitted as an update to Fedora 31. https://bodhi.fedoraproject.org/updates/FEDORA-2019-3d6348341f
Doesn't timedatex need to be removed from comps?
yes, and also, timedatex should be retired, if it's obsolete. It is not yet retired from F31 or F32. I'm sending PRs for comps. This has broken Server install on Rawhide.
https://pagure.io/fedora-comps/pull-request/413
systemd-243-2.gitfab6f01.fc31 has been pushed to the Fedora 31 testing repository. If problems still persist, please make note of it in this bug report. See https://fedoraproject.org/wiki/QA:Updates_Testing for instructions on how to install test updates. You can provide feedback for this update here: https://bodhi.fedoraproject.org/updates/FEDORA-2019-3d6348341f
timedatex has been retired in the F31 and master branches.
systemd-243-2.gitfab6f01.fc31 has been pushed to the Fedora 31 stable repository. If problems still persist, please make note of it in this bug report.