Time problem

[ptit...@gmail.com]

Hello,

I've searched in this group, and could find some answers, and tried some things, but still it doesn't work for me.
Maybe I am doing something wrong.

So here is the situation. I've installed Qubes today for the 1st time. That's version 4.

My laptop is configured in the BIOS with the time in UTC, and it says 21:11.

After I boot, my clock in the GUI says 19:11.
I go to dom0, and here is what I get:

Local Time: 19:11
UTC : 17:11
Time Zone: CEST +0200

I go to "sys-net" and I get

Local Time: 19:11
UTC : 17:11
Time Zone: CEST +0200

What I am expecting is

Local Time: 21:11
UTC : 19:11
Time Zone: CEST +0200


It seems that there is a confusion in the system between my laptop which is in UTC, and the system who thinks that the laptop time is local, and consider that UTC is -2 hours.

I've tried to change the clock with hwclock -w --utc, change with the command timedatectl.
I've checked if the service "systemd-timesyncd" is running on "sys-net", and it doesn't. I have this error message (output from systemctl status timesyncd.service"

● systemd-timesyncd.service - Network Time Synchronization
Loaded: loaded (/usr/lib/systemd/system/systemd-timesyncd.service; enabled; vendor preset: enabled)
Drop-In: /usr/lib/systemd/system/systemd-timesyncd.service.d
└─30_qubes.conf
Active: failed (Result: exit-code) since Sun 2018-09-16 19:06:54 CEST; 9min ago
Docs: man:systemd-timesyncd.service(8)
Process: 379 ExecStart=/usr/lib/systemd/systemd-timesyncd (code=exited, status=238/STATE_DIRECTORY)
Main PID: 379 (code=exited, status=238/STATE_DIRECTORY)

Sep 16 19:06:54 sys-net systemd[1]: systemd-timesyncd.service: Service has no hold-off time, scheduling restart.
Sep 16 19:06:54 sys-net systemd[1]: systemd-timesyncd.service: Scheduled restart job, restart counter is at 5.
Sep 16 19:06:54 sys-net systemd[1]: Stopped Network Time Synchronization.
Sep 16 19:06:54 sys-net systemd[1]: systemd-timesyncd.service: Start request repeated too quickly.
Sep 16 19:06:54 sys-net systemd[1]: systemd-timesyncd.service: Failed with result 'exit-code'.
Sep 16 19:06:54 sys-net systemd[1]: Failed to start Network Time Synchronization.

with journalctl -xe I get this error:

Sep 16 19:06:54 sys-net systemd[365]: systemd-timesyncd.service: Failed to set up special execution directory in /var/lib: File exists
Sep 16 19:06:54 sys-net systemd[365]: systemd-timesyncd.service: Failed at step STATE_DIRECTORY spawning /usr/lib/systemd/systemd-timesyncd: File exists
Sep 16 19:06:54 sys-net systemd[1]: systemd-timesyncd.service: Main process exited, code=exited, status=238/STATE_DIRECTORY
Sep 16 19:06:54 sys-net systemd[1]: systemd-timesyncd.service: Failed with result 'exit-code'.
Sep 16 19:06:54 sys-net systemd[1]: Failed to start Network Time Synchronization.
-- Subject: Unit systemd-timesyncd.service has failed
-- Defined-By: systemd
-- Support: https://lists.freedesktop.org/mailman/listinfo/systemd-devel
--
-- Unit systemd-timesyncd.service has failed.
--
-- The result is RESULT.



The only other thing I've done on my Qubes, is to upgrade everything to Fedora 28 instead of the base installation of Fedora 26. I mean "sys-net" is configured with Fedora 28 now.
I don't know if that could cause such a problem.

I will try to go back to Fedora 26. if that changes anything.

[ptit...@gmail.com]

Ok, after reassigning the template "federa-26" to "sys-net", then the time is ok.
And the systemd-timesyncd.service is running this time without errors.

But dom0 is not ok, the time is still wrong. It is also in Fedora 28, maybe I should go back to fedora 26...

I've followed how to upgrade to fedora 28. Maybe I've done something wrong.

Is there something that needs to be aware when doing that upgrade?
I would like to avoid to stay on Fedora 26.

[ptit...@gmail.com]

After going back to many of the changes I've done, the time in dm0 is always wrong.
All the other VM have the correct time. I don't know what to do...

the hwclock output in sys-net and dm0 are not the same. always the 2 hours of difference.

[Ivan Mitev]

This might help:

https://github.com/Qubes-Community/Contents/blob/master/docs/system/clock-time.md

[ptit...@gmail.com]

Thank you.
After using Qubes for other things, I've seen that the time in dom0 was corrected automatically.
So by going back to Fedora 26 in sys-net solved my problem.
Maybe I should open an issue ticket.

Nevertheless I will check it with your link, seems exactly what I needed yesterday! :-D

[awokd]

ptit...@gmail.com:

https://github.com/QubesOS/qubes-issues/issues/3983 - Fedora 28 should
be fixed in a few days once the patch gets pushed from Testing to
general release.