| Summary: | Session restore with no apps restored is much slower than new session mode | ||
|---|---|---|---|
| Product: | [Plasma] plasmashell | Reporter: | medin <med.medin.2014> |
| Component: | Session Management | Assignee: | Plasma Bugs List <plasma-bugs-null> |
| Status: | RESOLVED FIXED | ||
| Severity: | normal | CC: | kdedev, nate, plasma-bugs-null |
| Priority: | NOR | ||
| Version First Reported In: | 5.21.4 | ||
| Target Milestone: | 1.0 | ||
| Platform: | Other | ||
| OS: | Linux | ||
| Latest Commit: | Version Fixed/Implemented In: | ||
| Sentry Crash Report: | |||
| Attachments: | log | ||
|
Description
medin
2021-04-23 00:14:35 UTC
Please confirm this happens for a new user. Please also provide a dbus-monitor --pcap log of the boot processs (In reply to David Edmundson from comment #1) > Please confirm this happens for a new user. > > Please also provide a dbus-monitor --pcap log of the boot processs With new user both modes take the same time to boot to desktop, I think my current user is bloated somewhere, I don't know why it takes so much time to boot, maybe due to having some Akonadi apps like KAlarm, KMail (configured with GMail), KNotes. I did run dbus-monitor --pcap but it appends terminal output as long as I use other apps, I don't know how to run it correctly. `dbus-monitor --pcap > /tmp/someLogFile &` and run super early. Both being slow makes some sense at least - we wait for apps to finish starting up if autostarted. We have had situations where autostart apps would claim to support statup notifications and then do a bad job of implementing them. going through the relevant autostarted app would help find that Here's the result of dbus-monitor --pcap Created attachment 137871 [details]
log
Thanks for the bug report. I'm sorry we weren't able to get to this yet. There have been many fixes and improvements in Plasma 6, and this issue may have been fixed. Can you please re-test on your system with Plasma 6.3.4 or later and let us know if you can still reproduce the problem? Thanks! (In reply to TraceyC from comment #6) > Thanks for the bug report. I'm sorry we weren't able to get to this yet. > There have been many fixes and improvements in Plasma 6, and this issue may > have been fixed. > > Can you please re-test on your system with Plasma 6.3.4 or later and let us > know if you can still reproduce the problem? Thanks! I retested both modes "On last logout" and "Start with an empty session" with reboot and logout, and measured the time with my phone stopwatch from SDDM to when it fully shows desktop icons, the results showed that both modes have quite similar consumed time. The only note is that with reboot, both modes consumes a time x5 than what is needed for logging from a logout. |