Bug 1510301

Proposing as final blocker because it violates Final Release Requirements/Post-install requirements/System services (http://fedoraproject.org/wiki/Fedora_27_Final_Release_Criteria#System_services).

May depend on how fast you login again, see /etc/pulse/daemon.conf option:
; exit-idle-time = 20

which means, pulseaudio may stick around for at least 20 seconds after logout.

You can consider setting that shorter, and/or set
KillUserProcesses=yes
in /etc/systemd/logind.conf

For what it's worth, and given comment #2 and my experience, this is not new and not a regression in behavior.  I'm -1 and against treating this as any sort of blocker

I think this is related to the issues around https://fedoraproject.org/wiki/Changes/KillUserProcesses_by_default, which is not an easy problem. I don't think this violates the release criteria.

Same problem after update on Fedora 26.

(In reply to Rex Dieter from comment #2)
> May depend on how fast you login again, see /etc/pulse/daemon.conf option:
> ; exit-idle-time = 20
> 
> which means, pulseaudio may stick around for at least 20 seconds after
> logout.
> 
> You can consider setting that shorter, and/or set
> KillUserProcesses=yes
> in /etc/systemd/logind.conf

Works as you have said with exit-idle-time reduced; need to wait at least that amount of time before logging in again. PulseAudio restarts properly with exit-idle-time set to 0.

-1 blocker. I don't believe this violates the criterion cited, which is: "All system services present after installation with one of the release-blocking package sets must start properly, unless they require hardware which is not present". For a start, pulseaudio is not a system service. For a second, in basic usage - which is really what the criterion covers - it *does* start correctly. It only fails in this more complex scenario (logging out and back in again).

-1 blocker

I was able to reproduce this with KDE. But I don't think that we should block on this. I agree with Adam that to break it user has to do more steps. This could problem on multi-user systems (like in schools). But I don't think that running sounds is the main feature on such systems.

There is quite an easy workaround:
pkill pulseaudio; pulseaudio --start

-1 blocker; +1 FE in case that we won't ship this week.

I'm going to disagree with others here, but I consider this to be a blocker. Logging out and back in is not a complex scenario, but a pretty basic one (and covered in criteria). Sound is a basic functionality, also covered in criteria. 

If this was discovered earlier, I'd be +1 blocker. However, since this was discovered very late in the process, and this problem, while inconvenient, is not  really critical (should not be encountered too often, has no serious consequences), I'm fine with rejecting this and (hopefully) fix this with a later update.

I'd like to have this accepted as a F28 blocker, though. Proposing right away.

Kamil -- note that it's not just something that happens every time you log out. It's something that happens if you log out and log back in again very quickly. 

I'm -1 on it as a blocker for F28, too, unless we're going to add whole classes of annoying bugs to the potential-blocker category. And for that matter, this appears to be something that could be solved by a zero day (sure, one could potentially recreate in a live session, but... generally one wouldn't).

Discussed at 2017-11-09 Final Go/No-Go meeting, acting as a blocker review meeting: https://meetbot-raw.fedoraproject.org/fedora-meeting-1/2017-11-09/f27-final-and-server-beta-go-no-go-meeting.2017-11-09-18.02.html . This was rejected as a blocker for both F27 and F28; there was general agreement that it doesn't really constitute a violation of any criteria and is only an annoyance that's not severe enough to block the release. We will document the issue in commonbugs and advise waiting 20 seconds between logins, or changing the setting.

*** Bug 1517475 has been marked as a duplicate of this bug. ***

%changelog
* Mon Jan 08 2018 Rex Dieter <rdieter> - 11.1-8
- exit-idle-time = 4 (#1510301)

Reduce default idle time (20 -> 4) to mitigate this

pulseaudio-11.1-13.fc27 has been submitted as an update to Fedora 27. https://bodhi.fedoraproject.org/updates/FEDORA-2018-ef0b897e5d

pulseaudio-11.1-13.fc27 has been pushed to the Fedora 27 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-2018-ef0b897e5d

pulseaudio-11.1-15.fc27 has been submitted as an update to Fedora 27. https://bodhi.fedoraproject.org/updates/FEDORA-2018-ef0b897e5d

pulseaudio-11.1-15.fc27 has been pushed to the Fedora 27 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-2018-ef0b897e5d

*** Bug 1554125 has been marked as a duplicate of this bug. ***

pulseaudio-11.1-15.fc27 has been pushed to the Fedora 27 stable repository. If problems still persist, please make note of it in this bug report.