#1104850 .remove-on-upgrade /etc/NetworkManager/dispatcher.d/ntpsec-ntpdate: Not found in archive

Package:
network-manager
Source:
network-manager
Description:
network management framework (daemon and userspace tools)
Submitter:
Francesco Potortì
Date:
2026-02-10 09:41:01 UTC
Severity:
normal
#1104850#5
Date:
2025-05-07 09:47:50 UTC
From:
To:
I get these every day on syslog:

apt.systemd.daily[1016180]: tar: .remove-on-upgrade /etc/NetworkManager/dispatcher.d/ntpsec-ntpdate: Not found in archive
apt.systemd.daily[1016180]: tar: Exiting with failure status due to previous errors

#1104850#10
Date:
2025-09-15 13:42:48 UTC
From:
To:
Control: reassign -1 ntpsec
This hook is not shipped by network-manager.
Most likely it's shipped by ntpsec, thus re-assigning

Michael

#1104850#19
Date:
2026-02-08 05:39:40 UTC
From:
To:
Michael,

I finally got a chance to look at this.

I don't really understand this at all. But most importantly, I don't see
what I could do differently. NTPsec has very basic remove-on-upgrade for
some old hooks in debian/<package>.conffiles.

Do you have any ideas? This change came from you, actually:
https://salsa.debian.org/debian/ntpsec/-/commit/2442380eebc2a401ecc5d1c74f3dc6f71bc37943

That change was released in bookworm, which is now oldstable, so I can
probably just remove those remove-on-upgrade entries at this time.

#1104850#24
Date:
2026-02-08 13:05:36 UTC
From:
To:
The log looks like the user is running some form of backup thing in
the apt.systemd.daily job, that looks at conffile registrations and
backs them up. Maybe this is even an apt feature.
However this thing doesn't seem to understand the .remove-on-upgrade
feature.

Would be sensible to figure out what this actually is and reassign
there.

Chris

#1104850#29
Date:
2026-02-08 13:10:42 UTC
From:
To:
* Chris Hofstaedtler <zeha@debian.org> [260208 14:07]:

It'd appear this is unattended-upgrades.

Chris

#1104850#40
Date:
2026-02-10 09:39:11 UTC
From:
To:
Hi Richard, hi Chris!

Am 08.02.26 um 06:39 schrieb Richard Laager:

When (re)assigning this to nptsec, I should have checked this more
carefully. I agree, that what ntpsec is doing looks fine and the bug is
elsewhere.
Thanks for investigating that further, Chris (re. unattended-upgrades).

That said, yes, I think you can safely drop this migration code now.


Regards,
Michael