KDE Plasma spectacle fails to start after some uptime, dbus is broken?

good.dove4289 · 2025-07-18 02:33:51

SUMMARY

spectacle fails to start after some uptime. The spectacle process hangs, and gdb shows it is waiting in the following stack trace:

see https://bugs.kde.org/show_bug.cgi?id=505400 for stack trace

Rarely, after waiting anywhere between a few seconds and several minutes, the screenshot UI shows up.

The following shows up in systemd journal:

spectacle[3180927]: Error querying plasma version "org.freedesktop.DBus.Error.NoReply" "Did not receive a reply. Possible causes include: the remote application did not send a reply, the message bus security policy blocked the reply, the reply timeout expired, or the network connection was broken."

STEPS TO REPRODUCE
1. Use Plasma for a while (not sure what the exact cause is)
2. Launch spectacle to take screenshot

❯ pacman -Q spectacle
spectacle 1:6.4.2-2

Additional information:

After this starts happening, other applications show similar errors in journal:

chromium[2434]: [2434:2434:0717/193131.160910:ERROR:dbus/object_proxy.cc:590] Failed to call method: org.freedesktop.Notifications.Notify: object_path= /org/freedesktop/Notifications: org.freedesktop.DBus.Error.NoReply: Did not receive a reply. Possible causes include: the remote application did not send a reply, the message bus security policy blocked the reply, the reply timeout expired, or the network connection was broken.

skunktrader · 2025-07-18 02:45:46

Is knotify (or some other notification daemon) running?

good.dove4289 · 2025-07-18 02:50:24

"knotify" didn't match any lines in journalctl, and can't find any process "knotify" running, but I can receive notifications. (notify-send works, or at least it did before having dbus issues)

Last edited by good.dove4289 (2025-07-18 02:50:49)

seth · 2025-07-18 05:11:24

hostnamectl
loginctl session-status
ps aux | grep dbus

good.dove4289 · 2025-07-18 05:15:54

https://hst.sh/raw/edazekefis

Note that this is after rebooting, problem is gone for now

Problem only starts after a few days of uptime, can't find a specific way to reproduce

Last edited by good.dove4289 (2025-07-18 05:19:24)

seth · 2025-07-18 06:08:13

 Static hostname:

https://wiki.archlinux.org/title/Networ … e_hostname
dbus is network sensitive, if your (dynamic) hostname changes after it started (eg. from NM because of a re-lease) you'll lose the connection.

good.dove4289 · 2025-07-18 06:14:08

I changed the hostname a few times with `hostnamectl`, the problem is not occuring.

seth · 2025-07-18 06:24:27

This is not how this works. "hostnamectl hostname foo" sets the static hostname.
You could change the hostname using [cdrumroll] "hostname" and even then might not immediately just run into it (this will only apply when new clients try to use the new hostname while the session bus is running on the old one

The problem is frequent (as frequent as people skipping the wiki on installation)
Set a static hostname, reboot, wait for the problem to re-emerge.
If that doesn't happen and please always remember to mark resolved threads by editing your initial posts subject - so others will know that there's no task left, but maybe a solution to find.
Thanks.

good.dove4289 · 2025-07-20 18:44:27

Issue is back, received error

spectacle[1135215]: Error querying plasma version "org.freedesktop.DBus.Error.NoReply" "Did not receive a reply. Possible causes include: the remote application did not send a reply, the message bus security policy blocked the reply, the reply timeout expired, or the network connection was broken."

Not sure if related, but this started after compiling chromium overnight.

good.dove4289 · 2025-07-20 19:08:37

Could this be related?

https://hst.sh/awadaqujuk.apache

I was using scx scheduler and some task stalled

seth · 2025-07-20 21:39:25

https://hst.sh/awadaqujuk.apache are some largely isolated error messages?

Not sure if related, but this started after compiling chromium overnight.

Did you run OOM and the oom killer started to kill some processes - incl. plasma or dbus?

With less than 32GB RAM (and no swap), don't even try to build chromium.

good.dove4289 · 2025-07-20 21:47:07

Journal doesn't have any oom messages, I have 96GB of memory so it shouldn't be an issue.

seth · 2025-07-20 21:56:50

'key, so what does the journal do have?
Please post your complete system journal for the relevant boot, eg.

sudo journalctl -b -1 | curl -F 'file=@-' 0x0.st

for the previous (-1) one.

Arch Linux

#1 2025-07-18 02:33:51

KDE Plasma spectacle fails to start after some uptime, dbus is broken?

#2 2025-07-18 02:45:46

Re: KDE Plasma spectacle fails to start after some uptime, dbus is broken?

#3 2025-07-18 02:50:24

Re: KDE Plasma spectacle fails to start after some uptime, dbus is broken?

#4 2025-07-18 05:11:24

Re: KDE Plasma spectacle fails to start after some uptime, dbus is broken?

#5 2025-07-18 05:15:54

Re: KDE Plasma spectacle fails to start after some uptime, dbus is broken?

#6 2025-07-18 06:08:13

Re: KDE Plasma spectacle fails to start after some uptime, dbus is broken?

#7 2025-07-18 06:14:08

Re: KDE Plasma spectacle fails to start after some uptime, dbus is broken?

#8 2025-07-18 06:24:27

Re: KDE Plasma spectacle fails to start after some uptime, dbus is broken?

#9 2025-07-20 18:44:27

Re: KDE Plasma spectacle fails to start after some uptime, dbus is broken?

#10 2025-07-20 19:08:37

Re: KDE Plasma spectacle fails to start after some uptime, dbus is broken?

#11 2025-07-20 21:39:25

Re: KDE Plasma spectacle fails to start after some uptime, dbus is broken?

#12 2025-07-20 21:47:07

Re: KDE Plasma spectacle fails to start after some uptime, dbus is broken?

#13 2025-07-20 21:56:50

Re: KDE Plasma spectacle fails to start after some uptime, dbus is broken?

Board footer