I noticed since yesterday, that the icon, which indicates that updates are available, never gets cleared on my system, although I obviously try to launch the updater in a timely fashion - and - the operation succeeds …
It happened to me as well and is most probably an issue with the pulseaudio.
Does it happen with fedora-32 template only or is it the case for other templates too?
If only with fedora-32, try to directly open the fedora-32 template and execute:
$ sudo dnf update
If it says anything about pulseaudio, try to execute:
$ sudo dnf update --best --allowerasing
This should solve the issue.
I have the same persistent notification due to pulseaudio updates. Do you recall what packages it said it will remove when you run with --best --allowerasing.
I am told it will remove
Removing dependent packages:
pulseaudio-qubes
qubes-vm-recommended
Which at least the second one sounds important for qubes operation so feeling a bit cautious
Worked for me! Update sign is gone at the taskbar.
Thanks a bunch Victor.
Here’s my output:
[user@fedora-32 ~]$ sudo dnf update --best --allowerasing
Last metadata expiration check: 0:11:37 ago on Sat Dec 5 20:32:06 2020.
Dependencies resolved.
Worked for me! Update sign is gone at the taskbar.
Thanks a bunch Victor.
I did NOT yet perform the second part of the recommended procedure from ‘taran1s’ …
Here’s my output:
[user@fedora-32 ~]$ sudo dnf update --best --allowerasing
Last metadata expiration check: 0:11:37 ago on Sat Dec 5 20:32:06 2020.
Dependencies resolved.
I’d still like to get a confirmation from someone closer to the source, that this is all expected / OK - and - there’s nothing to worry about. - TIA again …
I noticed since yesterday, that the icon, which indicates that updates are
available, never gets cleared on my system, although I obviously try to
launch the updater in a timely fashion - and - the operation succeeds …
Here’s the log from the latest attempt:
Updating fedora-32
fedora-32:
ID: dnf list updates --refresh >/dev/null
Function: cmd.run
Result: True
Comment: Command “dnf list updates --refresh >/dev/null” run
Started: 09:00:59.753451
Duration: 8745.114 ms
Changes:
pid:
1077
retcode:
0
stderr:
stdout:
ID: update
Function: pkg.uptodate
Result: True
Comment: Upgrade ran successfully
Started: 09:01:10.612928
Duration: 24382.315 ms
Changes:
ID: notify-updates
Function: cmd.run
Name: /usr/lib/qubes/upgrades-status-notify
Result: True
Comment: Command “/usr/lib/qubes/upgrades-status-notify” run
Started: 09:01:34.995429
Duration: 3878.256 ms
Changes:
pid:
1148
retcode:
0
stderr:
stdout:
Summary for fedora-32
Succeeded: 3 (changed=2)
Failed: 0
Total states run: 3
Total run time: 37.006 s
Does anyone have an explanation - or - a suggestion what else to try? - TIA!
In which case, it’s a known bug, and the fix is in current-testing.
Thanks for the link to this issue. - I’ll spend some time tomorrow reading it in detail & will report back here …
Yes, it looks like that this is / was most likely the reason for the ‘hickup’ on my system as well.
In the meantime the situation has changed however on my system again!
As I did report already initially I update my system in a timely fashion - and - continued to do so, even while waiting for an anwer / response related to “sudo dnf update --best --allowerasing” …
Today two more updates related to ‘fedora-32’ and ‘whonix-gw-15’ were performed …
Result: Update of template ‘whonix-gw-15’ succeeded immediately - and - the partial update of ‘fedora-32’ succeeded after multiple ‘hickups’. - However the Qubes Updater icon is still active - but - when I perform ‘sudo dnf update’ in Dom0 the system now only reports:
[vr@dom0 ~]$ sudo dnf update
Qubes OS Repository for Dom0 80 MB/s | 82 kB 00:00
Dependencies resolved.
Nothing to do.
Complete!
[vr@dom0 ~]$
while running the update of the ‘fedora-32’ template using the updater still reports: