Bug 466634

Summary: When the desktop is running, even at idle, plasmashell will consume all of my 32GB of memory resulting in a system freeze.
Product: [Plasma] plasmashell Reporter: kevin.david.hall
Component: generic-performanceAssignee: Plasma Bugs List <plasma-bugs>
Status: RESOLVED DUPLICATE    
Severity: crash CC: nate
Priority: NOR Keywords: regression
Version: 5.27.2   
Target Milestone: 1.0   
Platform: Arch Linux   
OS: Linux   
Latest Commit: Version Fixed In:
Sentry Crash Report:
Attachments: plasmsashell memory usage after 1 hour 40 minutes of uptime
plasmashell after sitting idle overnight, almost all memory saturated
display setup.jpeg
kscreen_console_bug_output.txt
attachment-3365820-0.html
attachment-3374239-0.html
attachment-3375705-0.html
attachment-3387160-0.html
attachment-3493340-0.html
attachment-3857887-0.html

Description kevin.david.hall 2023-03-01 05:32:29 UTC
Created attachment 156848 [details]
plasmsashell memory usage after 1 hour 40 minutes of uptime

SUMMARY
SYSTEM INFO.
Operating System: Arch Linux 
KDE Plasma Version: 5.27.2
KDE Frameworks Version: 5.103.0
Qt Version: 5.15.8
Kernel Version: 6.2.1-arch1-1 (64-bit)
Graphics Platform: Wayland
Processors: 16 × AMD Ryzen 7 5800X 8-Core Processor
Memory: 31.3 GiB of RAM
Graphics Processor: AMD Radeon RX 6600 XT


***
NOTE: If you are reporting a crash, please try to attach a backtrace with debug symbols.
See https://community.kde.org/Guidelines_and_HOWTOs/Debugging/How_to_create_useful_crash_reports
***


STEPS TO REPRODUCE
1. Boot computer and log into a desktop session (same result if using X11 or Wayland)
2. Wait...initial plasmashell memory footprint is ~2GB, over the course of several hours, it will grow occupy all memory.

OBSERVED RESULT
Computer hard locks and requires a reboot.

EXPECTED RESULT
Computer runs normally. 

SOFTWARE/OS VERSIONS
Windows: 
macOS: 
Linux/KDE Plasma: 
(available in About System)
KDE Plasma Version: 5.27.2
KDE Frameworks Version: 5.103.0
Qt Version: 5.15.8

ADDITIONAL INFORMATION
This behavior was first observed ~18 hours ago after a system update. Before that, I could go weeks without issue. System is fully up to date as of this report.
Comment 1 kevin.david.hall 2023-03-01 13:38:28 UTC
Created attachment 156870 [details]
plasmashell after sitting idle overnight, almost all memory saturated
Comment 2 Nate Graham 2023-03-01 20:48:45 UTC
Do you have multiple screens, or only one?
Comment 3 kevin.david.hall 2023-03-01 21:03:33 UTC
Created attachment 156899 [details]
display setup.jpeg

Multiple displays (2). Attached is shot of current display setup in system
settings.

On Wed, Mar 1, 2023 at 1:48 PM Nate Graham <bugzilla_noreply@kde.org> wrote:

> https://bugs.kde.org/show_bug.cgi?id=466634
>
> Nate Graham <nate@kde.org> changed:
>
>            What    |Removed                     |Added
>
> ----------------------------------------------------------------------------
>              Status|REPORTED                    |NEEDSINFO
>          Resolution|---                         |WAITINGFORINFO
>                  CC|                            |nate@kde.org
>
> --- Comment #2 from Nate Graham <nate@kde.org> ---
> Do you have multiple screens, or only one?
>
> --
> You are receiving this mail because:
> You reported the bug.
Comment 4 Nate Graham 2023-03-01 21:36:32 UTC
Hmm, reminds me of Bug 466362. Does it stop happening if you only have one screen plugged in? If so, does it start happening again when you plug in the second screen?

Also can you run `kscreen-console bug` and paste the "EDID" sections for your screens?

Thanks!
Comment 5 kevin.david.hall 2023-03-01 21:39:58 UTC
Created attachment 156900 [details]
kscreen_console_bug_output.txt

Put it in a txt file since it's so long.
I haven't tried it with only one screen. Does it need to be unplugged, or
will powered down suffice?

On Wed, Mar 1, 2023 at 2:36 PM Nate Graham <bugzilla_noreply@kde.org> wrote:

> https://bugs.kde.org/show_bug.cgi?id=466634
>
> --- Comment #4 from Nate Graham <nate@kde.org> ---
> Hmm, reminds me of Bug 466362. Does it stop happening if you only have one
> screen plugged in? If so, does it start happening again when you plug in
> the
> second screen?
>
> Also can you run `kscreen-console bug` and paste the "EDID" sections for
> your
> screens?
>
> Thanks!
>
> --
> You are receiving this mail because:
> You reported the bug.
Comment 6 Nate Graham 2023-03-01 21:43:19 UTC
Unplugged would be better, since it's DisplayPort.

Looks like your two screens report identical serial numbers. Despite thir, their EDID hash values do differ, so that's nice at least.

Definitely curious to know if the issue goes away with only one screen plugged in.
Comment 7 kevin.david.hall 2023-03-01 21:45:37 UTC
Created attachment 156901 [details]
attachment-3365820-0.html

Okay, I'll pull one monitor. plasmashell is currently sitting at 20.3Gb
memory usage as of this email. I'll report back in a couple of hours as to
the change or lack thereof.

On Wed, Mar 1, 2023 at 2:43 PM Nate Graham <bugzilla_noreply@kde.org> wrote:

> https://bugs.kde.org/show_bug.cgi?id=466634
>
> --- Comment #6 from Nate Graham <nate@kde.org> ---
> Unplugged would be better, since it's DisplayPort.
>
> Looks like your two screens report identical serial numbers. Despite thir,
> their EDID hash values do differ, so that's nice at least.
>
> Definitely curious to know if the issue goes away with only one screen
> plugged
> in.
>
> --
> You are receiving this mail because:
> You reported the bug.
Comment 8 kevin.david.hall 2023-03-01 23:20:36 UTC
Created attachment 156902 [details]
attachment-3374239-0.html

plasmashell got to about 23Gb in size and crashed/restarted. When it
respawned, it grabbed ~390Mb of memory instead of the 2Gb that I was seeing
earlier on startup. Still have the one monitor removed from desktop,
running in single monitor since the last email.

On Wed, Mar 1, 2023 at 2:45 PM Kevin Hall <kevin.david.hall@gmail.com>
wrote:

> Okay, I'll pull one monitor. plasmashell is currently sitting at 20.3Gb
> memory usage as of this email. I'll report back in a couple of hours as to
> the change or lack thereof.
>
> On Wed, Mar 1, 2023 at 2:43 PM Nate Graham <bugzilla_noreply@kde.org>
> wrote:
>
>> https://bugs.kde.org/show_bug.cgi?id=466634
>>
>> --- Comment #6 from Nate Graham <nate@kde.org> ---
>> Unplugged would be better, since it's DisplayPort.
>>
>> Looks like your two screens report identical serial numbers. Despite thir,
>> their EDID hash values do differ, so that's nice at least.
>>
>> Definitely curious to know if the issue goes away with only one screen
>> plugged
>> in.
>>
>> --
>> You are receiving this mail because:
>> You reported the bug.
>
>
Comment 9 Nate Graham 2023-03-01 23:24:04 UTC
Thanks. And if you plug in the second screen, does the memory use immediately skyrocket?
Comment 10 kevin.david.hall 2023-03-01 23:34:26 UTC
Created attachment 156903 [details]
attachment-3375705-0.html

It rose to ~400Mb. I'll keep an eye on it and let you know if it climbs
over time.

On Wed, Mar 1, 2023 at 4:24 PM Nate Graham <bugzilla_noreply@kde.org> wrote:

> https://bugs.kde.org/show_bug.cgi?id=466634
>
> --- Comment #9 from Nate Graham <nate@kde.org> ---
> Thanks. And if you plug in the second screen, does the memory use
> immediately
> skyrocket?
>
> --
> You are receiving this mail because:
> You reported the bug.
Comment 11 kevin.david.hall 2023-03-02 02:20:04 UTC
Created attachment 156906 [details]
attachment-3387160-0.html

It's been almost 3 hours. Still hovering in the 400Mb of memory range. So,
is there a permanent fix, or do I need to start my machine with 1 monitor
for the foreseeable future?

On Wed, Mar 1, 2023 at 4:34 PM Kevin Hall <kevin.david.hall@gmail.com>
wrote:

> It rose to ~400Mb. I'll keep an eye on it and let you know if it climbs
> over time.
>
> On Wed, Mar 1, 2023 at 4:24 PM Nate Graham <bugzilla_noreply@kde.org>
> wrote:
>
>> https://bugs.kde.org/show_bug.cgi?id=466634
>>
>> --- Comment #9 from Nate Graham <nate@kde.org> ---
>> Thanks. And if you plug in the second screen, does the memory use
>> immediately
>> skyrocket?
>>
>> --
>> You are receiving this mail because:
>> You reported the bug.
>
>
Comment 12 Nate Graham 2023-03-02 04:39:10 UTC
Definitely not a fix, just a workaround that helps to confirm that this is Bug 466362. We'll see if we can get it fixed soon!
Comment 13 Nate Graham 2023-03-02 04:39:25 UTC

*** This bug has been marked as a duplicate of bug 466362 ***
Comment 14 kevin.david.hall 2023-03-03 01:42:18 UTC
Created attachment 156942 [details]
attachment-3493340-0.html

Awesome Nate, thanks for the quick turnaround and attentiveness. I really
appreciate it. Wish the devs I work with were this quick.

On Wed, Mar 1, 2023 at 9:39 PM Nate Graham <bugzilla_noreply@kde.org> wrote:

> https://bugs.kde.org/show_bug.cgi?id=466634
>
> --- Comment #12 from Nate Graham <nate@kde.org> ---
> Definitely not a fix, just a workaround that helps to confirm that this is
> Bug
> 466362. We'll see if we can get it fixed soon!
>
> --
> You are receiving this mail because:
> You reported the bug.
Comment 15 Nate Graham 2023-03-06 16:47:36 UTC
Well the issue isn't fixed (just marked as a duplicate of another known issue) so don't thank me yet! :)
Comment 16 kevin.david.hall 2023-03-06 16:50:23 UTC
Created attachment 157049 [details]
attachment-3857887-0.html

I get it. Also, the stability did NOT survive a reboot (and I didn't expect
it to) but I did recreate the method of removing a monitor and rebooting as
instructed, but still getting over usage of memory even with a single
monitor.

On Mon, Mar 6, 2023 at 9:47 AM Nate Graham <bugzilla_noreply@kde.org> wrote:

> https://bugs.kde.org/show_bug.cgi?id=466634
>
> --- Comment #15 from Nate Graham <nate@kde.org> ---
> Well the issue isn't fixed (just marked as a duplicate of another known
> issue)
> so don't thank me yet! :)
>
> --
> You are receiving this mail because:
> You reported the bug.