Summary: | Become aware of offline updates and offer opt out | ||
---|---|---|---|
Product: | [Plasma] plasmashell | Reporter: | Harald Sitter <sitter> |
Component: | Theme - Breeze | Assignee: | Plasma Bugs List <plasma-bugs> |
Status: | RESOLVED DUPLICATE | ||
Severity: | wishlist | CC: | alex765, johnmaverick74, kde, nate, plasma-bugs, tobi.goergens, visual-design |
Priority: | HI | Keywords: | usability |
Version: | 5.27.5 | ||
Target Milestone: | 1.0 | ||
Platform: | Other | ||
OS: | Linux | ||
Latest Commit: | Version Fixed In: |
Description
Harald Sitter
2021-05-11 12:58:25 UTC
The logout screen itself lives in plasma, which is where the UI would need to be added. However kmserver is probably where the code to expose these features would need to be added, if they don't already exist. *** Bug 435971 has been marked as a duplicate of this bug. *** Couldn't you do it "just" in the logout greeter? I'm not sure what just means here. I'll also not pretend to have looked at either code base in years though ^^ Roughly we'll need: backendy bits talk to packagekit over dbus to figure out whether offline updates are staged (a), to set a different trigger type (c), to disable the offline trigger (b). Unrelated to packagekit we need to present a poweroff action that indeed behind the scenes triggers a reboot, not a poweroff (c). Ignoring re-staging after an opt-out for now, as IMO that had been better solved by discover anyway, if we can easily talk to PK from the greeter than I'd agree with this being doable there rather than ksmserver. Can we be a lot more sparing with the "VHI" flag. It has already become completely meaningless. In terms of logout order of events is: - user clicks button in kickoff this makes a DBus call - This DBus call auto-activates logout-greeter, this shows a prompt - On clicking confirm it makes a new DBus call - This DBus call auto-activates a helper binary called plasma-shutdown - This binary calls into ksmserver to ask apps to quit (which they can cancel) logout-greeter is the only visual one, though it's worth pointing out it can be skipped and the call in kickoff can go straight to plasma-shutdown based on user preferences. side note: If this isn't a VHI issue I don't know what is. The UX that this bug report tracks fixing is generating significant negativity among our users in bug reports and social media. We risk losing users over it. It may not be a crash or a bug per se, but the consequences to not improving the situation here would be just as severe IMO. (In reply to David Edmundson from comment #5) > Can we be a lot more sparing with the "VHI" flag. It has already become > completely meaningless. > > In terms of logout order of events is: > > - user clicks button in kickoff this makes a DBus call > - This DBus call auto-activates logout-greeter, this shows a prompt > > - On clicking confirm it makes a new DBus call > - This DBus call auto-activates a helper binary called plasma-shutdown > > - This binary calls into ksmserver to ask apps to quit (which they can > cancel) > > > logout-greeter is the only visual one, though it's worth pointing out it can > be skipped and the call in kickoff can go straight to plasma-shutdown based > on user preferences. Hi there, I'm working on a custom script to update my system at shutdown. The script itself is working, but it needs to be called before shutdown. I already tried to create a systemd script, but systemd shuts down services the script needs before it finishes, but I don't know which exactly are required. So, I had the idea to get the shutdown call by Plasma to redirect it to my script. Replacing the plasma-shutdown executable with my script worked, but it's also called when the system is rebooted. Is there any way to make the script just being called at shutdown? Which command is sent by plasma that actually shuts down the system? Do you know that? I'd be glad if you could help me here! What I've done so far can be found here: https://github.com/TobiPeterG/kubuntu_automatic_updates I'm looking into this and it seems feasible. Will try to fix this as well as Bug 435845. On second look, this is a duplicate of that, which is in the wrong place. *** This bug has been marked as a duplicate of bug 435845 *** |