|
Description
Jannis Liapis
2011-12-07 14:29:12 UTC
This is indeed the same as bug 270037, which is supposed to be fixed. Instead of reopening the older one, I keep this one open, because it has newer backtrace. *** Bug 297446 has been marked as a duplicate of this bug. *** *** Bug 300851 has been marked as a duplicate of this bug. *** Created attachment 73972 [details]
New crash information added by DrKonqi
kded4 ($Id$) on KDE Platform 4.8.4 (4.8.4) using Qt 4.8.1
- What I was doing when the application crashed:
The crash warning had already appeared when I opened my sleeping laptop.
-- Backtrace (Reduced):
#9 0x00007f2babe6d76d in QDBusInterfacePrivate::metacall (this=0x15d25c0, c=<optimized out>, id=0, argv=0x7fff61c8fb60) at qdbusinterface.cpp:278
#10 0x00007f2babe5937b in QDBusConnectionPrivate::deliverCall (this=0x14697b0, object=0x1535330, msg=..., metaTypes=..., slotIdx=5) at qdbusintegrator.cpp:947
#11 0x00007f2bae907446 in QObject::event (this=0x1535330, e=<optimized out>) at kernel/qobject.cpp:1195
#12 0x00007f2badc70894 in notify_helper (e=0x15e01c0, receiver=0x1535330, this=0x1296e60) at kernel/qapplication.cpp:4559
#13 QApplicationPrivate::notify_helper (this=0x1296e60, receiver=0x1535330, e=0x15e01c0) at kernel/qapplication.cpp:4531
*** Bug 307243 has been marked as a duplicate of this bug. *** Created attachment 74179 [details]
New crash information added by DrKonqi
kded4 ($Id$) on KDE Platform 4.8.5 (4.8.5) using Qt 4.8.1
- What I was doing when the application crashed:
I wake up the laptop, and then kde demon has crashed. but not very annoying
- Unusual behavior I noticed:
Some applications crashed sometimes after wake up like MysqlWorkbench and Synaptiks touchpad driver
-- Backtrace (Reduced):
#7 0xb3b2bd7c in Solid::Backends::UPower::Battery::slotChanged (this=0x8429200) at ../../../solid/solid/backends/upower/upowerbattery.cpp:130
[...]
#9 0xb3ac1395 in Solid::Backends::UPower::UPowerDevice::changed (this=0x84942a0) at moc_upowerdevice.cpp:105
#10 0xb3b2a1b0 in Solid::Backends::UPower::UPowerDevice::slotChanged (this=0x84942a0) at ../../../solid/solid/backends/upower/upowerdevice.cpp:219
#11 0xb3ac13cd in Solid::Backends::UPower::UPowerDevice::qt_static_metacall (_o=<optimized out>, _id=<optimized out>, _a=<optimized out>, _c=<optimized out>) at moc_upowerdevice.cpp:54
[...]
#13 0xb5b866a0 in QDBusInterfacePrivate::metacall (this=0x84db780, c=QMetaObject::InvokeMetaMethod, id=0, argv=0xbfd5be9c) at qdbusinterface.cpp:278
Created attachment 74227 [details]
New crash information added by DrKonqi
kded4 ($Id$) on KDE Platform 4.9.1 using Qt 4.8.3
- What I was doing when the application crash:
I pluged the poer cable to the laptop and resumed from "suspend to RAM".
The laptop was put to sleep while on battery.
-- Backtrace (Reduced):
#7 0xb3b03d08 in Solid::Backends::UPower::Battery::slotChanged (this=0x87d10e0) at ../../../solid/solid/backends/upower/upowerbattery.cpp:130
[...]
#9 0xb3aa00f5 in Solid::Backends::UPower::UPowerDevice::changed (this=0x8850e80) at moc_upowerdevice.cpp:105
#10 0xb3b021e3 in Solid::Backends::UPower::UPowerDevice::slotChanged (this=0x8850e80) at ../../../solid/solid/backends/upower/upowerdevice.cpp:219
#11 0xb3aa012d in Solid::Backends::UPower::UPowerDevice::qt_static_metacall (_o=<optimized out>, _id=<optimized out>, _a=<optimized out>, _c=<optimized out>) at moc_upowerdevice.cpp:54
[...]
#13 0xb5b5efe8 in QDBusInterfacePrivate::metacall (this=0x87ca0f8, c=QMetaObject::InvokeMetaMethod, id=0, argv=0xbf80ca9c) at qdbusinterface.cpp:278
Created attachment 74662 [details]
New crash information added by DrKonqi
kded4 ($Id$) on KDE Platform 4.8.5 (4.8.5) using Qt 4.8.1
- What I was doing when the application crashed:
I wake up the laptop (dell e6510), kde deamon crash.
-- Backtrace (Reduced):
#7 0xb3ab9d72 in Solid::Backends::UPower::Battery::slotChanged (this=0x91237f8) at ../../../solid/solid/backends/upower/upowerbattery.cpp:130
[...]
#9 0xb3a4f395 in Solid::Backends::UPower::UPowerDevice::changed (this=0x9123840) at moc_upowerdevice.cpp:105
#10 0xb3ab81b0 in Solid::Backends::UPower::UPowerDevice::slotChanged (this=0x9123840) at ../../../solid/solid/backends/upower/upowerdevice.cpp:219
#11 0xb3a4f3cd in Solid::Backends::UPower::UPowerDevice::qt_static_metacall (_o=<optimized out>, _id=<optimized out>, _a=<optimized out>, _c=<optimized out>) at moc_upowerdevice.cpp:54
[...]
#13 0xb5b146a0 in QDBusInterfacePrivate::metacall (this=0x9121588, c=QMetaObject::InvokeMetaMethod, id=0, argv=0xbf80508c) at qdbusinterface.cpp:278
Created attachment 75464 [details]
yet another crash report
My kded crashed with the same stacktrace (Solid::Backends::UPower::Battery::slotChanged).
Unusuall thing was that battery level was ~0% at wakeup (it seems to be the real status, is't increasing slowly now as I'm charging).
Git commit fb102411c4517cb9e62071ee92c6757d69ba3815 by Lukáš Tinkl. Committed on 27/11/2012 at 19:47. Pushed by lukas into branch 'KDE/4.10'. prevent crashes on suspend/resume when the battery vanishes by guarding the m_device pointer with QWeakPointer FIXED-IN: 4.9.4 M +10 -6 solid/solid/backends/upower/upoweracadapter.cpp M +27 -26 solid/solid/backends/upower/upowerbattery.cpp M +2 -1 solid/solid/backends/upower/upowerdeviceinterface.h M +3 -3 solid/solid/backends/upower/upowergenericinterface.cpp http://commits.kde.org/kdelibs/fb102411c4517cb9e62071ee92c6757d69ba3815 Git commit 44adb502ff0b78cdf90a49d0b6ba2f027bd6d3a7 by Lukáš Tinkl. Committed on 27/11/2012 at 19:47. Pushed by lukas into branch 'KDE/4.9'. prevent crashes on suspend/resume when the battery vanishes by guarding the m_device pointer with QWeakPointer FIXED-IN: 4.9.4 M +10 -6 solid/solid/backends/upower/upoweracadapter.cpp M +27 -26 solid/solid/backends/upower/upowerbattery.cpp M +2 -1 solid/solid/backends/upower/upowerdeviceinterface.h M +3 -3 solid/solid/backends/upower/upowergenericinterface.cpp http://commits.kde.org/kdelibs/44adb502ff0b78cdf90a49d0b6ba2f027bd6d3a7 *** Bug 315125 has been marked as a duplicate of this bug. *** *** Bug 316366 has been marked as a duplicate of this bug. *** So, apparently this is still not fixed. Anyone can reproduce this ? Created attachment 78138 [details]
New crash information added by DrKonqi
kded4 ($Id$) on KDE Platform 4.10.1 using Qt 4.8.3
- What I was doing when the application crashed:
Laptop was suspended due to inactivity.
Upon resumption, KDE daemon had crashed.
-- Backtrace (Reduced):
#6 Solid::Backends::UPower::Battery::slotChanged (this=0x1bf36f0) at ../../../solid/solid/backends/upower/upowerbattery.cpp:128
[...]
#9 0x00007fe3e05c2511 in QDBusInterfacePrivate::metacall (this=0x27f4a70, c=<optimized out>, id=0, argv=0x7fff6be511f0) at qdbusinterface.cpp:278
#10 0x00007fe3e05aded3 in QDBusConnectionPrivate::deliverCall (this=0x1ec0580, object=0x27ea7b0, msg=..., metaTypes=..., slotIdx=5) at qdbusintegrator.cpp:947
#11 0x00007fe3e324b46e in QObject::event (this=0x27ea7b0, e=<optimized out>) at kernel/qobject.cpp:1195
#12 0x00007fe3e25dde9c in QApplicationPrivate::notify_helper (this=this@entry=0x1cb83e0, receiver=receiver@entry=0x27ea7b0, e=e@entry=0x2792b90) at kernel/qapplication.cpp:4562
*** Bug 317171 has been marked as a duplicate of this bug. *** Could kdelibs 9187862f9caa2b2d41fcfef0637f3e57f27efdd5 have regressed this? Alex: yes, I got this crash on 22 March, with a 4.10 branch build. (see bug 317171 for full backtrace with debug symbols) The crash is on this line emit chargeStateChanged(m_chargeState, m_device.data()->udi()); I guess I could run kded in valgrind, at the expense of my CPU and productivity :) I wonder if reverting http://quickgit.kde.org/?p=kdelibs.git&a=commit&h=9187862f9caa2b2d41fcfef0637f3e57f27efdd5 might help. I'm not sure what the issue with losing signals is, mentioned in the commit. Maybe we can emit all the required signals in the "if (strong) { ... } else { emit }" path. But I'm not sure what signals would be needed - if I understand this right, the battery has been removed or gone missing. The battery was not removed during suspend to RAM but the AC adapter was disconnected. Created attachment 81169 [details]
New crash information added by DrKonqi
kded4 (4.10.5) on KDE Platform 4.10.5 using Qt 4.8.4
- What I was doing when the application crashed:
Open laptop lid to resume from sleep on Dell Latitude E6510, Fedora 19, KDE 4.10.5
-- Backtrace (Reduced):
#5 0x00000033652936a8 in Solid::Backends::UPower::Battery::slotChanged() () from /lib64/libsolid.so.4
[...]
#8 0x0000003bfbc36822 in QDBusInterfacePrivate::metacall(QMetaObject::Call, int, void**) () from /lib64/libQtDBus.so.4
#9 0x0000003bfbc24126 in QDBusConnectionPrivate::deliverCall(QObject*, int, QDBusMessage const&, QList<int> const&, int) () from /lib64/libQtDBus.so.4
#10 0x0000003bfa791e4e in QObject::event(QEvent*) () from /lib64/libQtCore.so.4
#11 0x0000003360fc846c in QApplicationPrivate::notify_helper(QObject*, QEvent*) () from /lib64/libQtGui.so.4
*** Bug 320175 has been marked as a duplicate of this bug. *** Created attachment 82049 [details]
New crash information added by DrKonqi
kded4 (4.10.5) on KDE Platform 4.10.5 using Qt 4.8.4
- What I was doing when the application crashed:
Pressed power button to resume from sleep (RAM hibernate) on Dell Latitude laptop.
-- Backtrace (Reduced):
#6 0x000000346d478411 in Solid::Backends::UPower::Battery::slotChanged (this=0x1aa9360) at /mnt/data/tmp/portage/kde-base/kdelibs-4.10.5-r1/work/kdelibs-4.10.5/solid/solid/backends/upower/upowerbattery.cpp:128
[...]
#9 0x0000003e4e42e8e9 in QDBusInterfacePrivate::metacall (this=0x1a52510, c=<optimized out>, id=0, argv=0x7fff06185e90) at qdbusinterface.cpp:278
#10 0x0000003e4e4208c7 in QDBusConnectionPrivate::deliverCall (this=0x165fbd0, object=0x1a98780, msg=..., metaTypes=..., slotIdx=5) at qdbusintegrator.cpp:951
#11 0x0000003e4e92a0f0 in QObject::event (this=0x1a98780, e=<optimized out>) at kernel/qobject.cpp:1194
#12 0x0000003e505da786 in notify_helper (e=0x1aa9fe0, receiver=0x1a98780, this=0x1498930) at kernel/qapplication.cpp:4562
(In reply to comment #20) > I wonder if reverting > http://quickgit.kde.org/?p=kdelibs. > git&a=commit&h=9187862f9caa2b2d41fcfef0637f3e57f27efdd5 might help. I'm not > sure what the issue with losing signals is, mentioned in the commit. > > Maybe we can emit all the required signals in the "if (strong) { ... } else > { emit }" path. But I'm not sure what signals would be needed - if I > understand this right, the battery has been removed or gone missing. Every time I resume from sleep, the battery manager claims I am out of battery right after I resume. This seems to suggest that KDE (or the kernel?) thinks the power is gone. Only shortly after it realizes it is actually there. So this other bug might also help trigger this bug. Regarding my bug report, AC adapter was disconnected. Created attachment 83080 [details]
New crash information added by DrKonqi
kded4 (4.11.2) on KDE Platform 4.11.2 using Qt 4.8.5
I had the same issue - a crash on resume. Battery or AC adaptier didn't change it's state during suspend/resume cycle.
-- Backtrace (Reduced):
#5 0x00007f47b0d5cc7e in Solid::Backends::UPower::Battery::slotChanged() () from /usr/lib64/libsolid.so.4
[...]
#8 0x00007f47b735b5d1 in QDBusInterfacePrivate::metacall(QMetaObject::Call, int, void**) () from /usr/lib64/qt4/libQtDBus.so.4
#9 0x00007f47b7347dc3 in QDBusConnectionPrivate::deliverCall(QObject*, int, QDBusMessage const&, QList<int> const&, int) () from /usr/lib64/qt4/libQtDBus.so.4
#10 0x00007f47ba5b9b0e in QObject::event(QEvent*) () from /usr/lib64/qt4/libQtCore.so.4
#11 0x00007f47b9751d5c in QApplicationPrivate::notify_helper(QObject*, QEvent*) () from /usr/lib64/qt4/libQtGui.so.4
Created attachment 83177 [details]
New crash information added by DrKonqi
kded4 (4.11.2) on KDE Platform 4.11.2 using Qt 4.8.5
- What I was doing when the application crashed:
Resume from sleeping (with battery marked as "faulty", in fact with a life reduced at 1 h but tolerable in my context)
-- Backtrace (Reduced):
#5 0x00000034bbaa0209 in Solid::Backends::UPower::Battery::slotChanged() () from /lib64/libsolid.so.4
[...]
#8 0x000000364da36822 in QDBusInterfacePrivate::metacall(QMetaObject::Call, int, void**) () from /lib64/libQtDBus.so.4
#9 0x000000364da24126 in QDBusConnectionPrivate::deliverCall(QObject*, int, QDBusMessage const&, QList<int> const&, int) () from /lib64/libQtDBus.so.4
#10 0x000000364d59232e in QObject::event(QEvent*) () from /lib64/libQtCore.so.4
#11 0x000000364edc84dc in QApplicationPrivate::notify_helper(QObject*, QEvent*) () from /lib64/libQtGui.so.4
Created attachment 83232 [details]
New crash information added by DrKonqi
kded4 (4.11.2) on KDE Platform 4.11.2 using Qt 4.8.5
- What I was doing when the application crashed:
Just opened the lid on the laptop and the crash was there. AC was connected and battery was not low on power
-- Backtrace (Reduced):
#5 0x00000032762a0297 in Solid::Backends::UPower::Battery::slotChanged() () from /lib64/libsolid.so.4
[...]
#8 0x0000003ce3836822 in QDBusInterfacePrivate::metacall(QMetaObject::Call, int, void**) () from /lib64/libQtDBus.so.4
#9 0x0000003ce3824126 in QDBusConnectionPrivate::deliverCall(QObject*, int, QDBusMessage const&, QList<int> const&, int) () from /lib64/libQtDBus.so.4
#10 0x0000003ce2f9232e in QObject::event(QEvent*) () from /lib64/libQtCore.so.4
#11 0x0000003ce47c84dc in QApplicationPrivate::notify_helper(QObject*, QEvent*) () from /lib64/libQtGui.so.4
*** Bug 327329 has been marked as a duplicate of this bug. *** update bug status *** Bug 327839 has been marked as a duplicate of this bug. *** *** Bug 328057 has been marked as a duplicate of this bug. *** *** Bug 328733 has been marked as a duplicate of this bug. *** *** Bug 329033 has been marked as a duplicate of this bug. *** Created attachment 84314 [details]
New crash information added by DrKonqi
kded4 (4.11.3) on KDE Platform 4.11.3 using Qt 4.8.4
- What I was doing when the application crashed:
Opened laptop lid to suspend from RAM (AC adapter was plugged in before, during, and after suspend to RAM).
Battery was 100% charged before S2RAM.
USB devices plugged in (before, during and after S2RAM):
Bus 001 Device 005: ID 7392:a822 Edimax Technology Co., Ltd (Edimax AC-1200 802.11ac WiFi adapter)
-- Backtrace (Reduced):
#6 0x00007ffd9745a6a7 in Solid::Backends::UPower::Battery::slotChanged (this=0x29cd270) at ../../../solid/solid/backends/upower/upowerbattery.cpp:149
[...]
#9 0x00007ffd9d9c8762 in QDBusInterfacePrivate::metacall (this=0x285e4d0, c=<optimized out>, id=0, argv=0x7fffe77516c0) at qdbusinterface.cpp:278
#10 0x00007ffd9d9b5e36 in QDBusConnectionPrivate::deliverCall (this=0x27a0320, object=0x2bd4470, msg=..., metaTypes=..., slotIdx=5) at qdbusintegrator.cpp:951
#11 0x00007ffda0648dce in QObject::event (this=0x2bd4470, e=<optimized out>) at kernel/qobject.cpp:1194
#12 0x00007ffd9f9cddfc in QApplicationPrivate::notify_helper (this=this@entry=0x2680130, receiver=receiver@entry=0x2bd4470, e=e@entry=0x2b92a40) at kernel/qapplication.cpp:4567
*** Bug 330156 has been marked as a duplicate of this bug. *** Created attachment 85185 [details]
New crash information added by DrKonqi
kded4 (4.12.2) on KDE Platform 4.12.2 using Qt 4.8.5
Power to laptop was lost and KDED4 crashed. This doesn't always happen but it does happen frequently.
-- Backtrace (Reduced):
#6 Solid::Backends::UPower::Battery::slotChanged (this=0x250eb80) at /usr/src/debug/kdelibs-4.12.2/solid/solid/backends/upower/upowerbattery.cpp:149
[...]
#9 0x00007f63e9a987f2 in QDBusInterfacePrivate::metacall (this=0x229a7d0, c=<optimized out>, id=0, argv=0x7fffc236ccd0) at qdbusinterface.cpp:278
#10 0x00007f63e9a85ead in QDBusConnectionPrivate::deliverCall (this=0x2077510, object=0x254f5e0, msg=..., metaTypes=..., slotIdx=5) at qdbusintegrator.cpp:951
#11 0x00007f63e971111e in QObject::event(QEvent*) () from /usr/lib64/libQtCore.so.4
#12 0x00007f63e8a968ac in QApplicationPrivate::notify_helper(QObject*, QEvent*) () from /usr/lib64/libQtGui.so.4
*** Bug 331779 has been marked as a duplicate of this bug. *** Created attachment 85452 [details]
New crash information added by DrKonqi
kded4 (4.12.3) on KDE Platform 4.12.3 using Qt 4.8.6
- What I was doing when the application crashed:
resume laptop from suspend to ram. also battery application in tray disappears and moves to desktop.
-- Backtrace (Reduced):
#6 Solid::Backends::UPower::Battery::slotChanged (this=0xe31fa0) at ../../../solid/solid/backends/upower/upowerbattery.cpp:149
[...]
#9 0x00007f5e74a91a82 in QDBusInterfacePrivate::metacall (this=0xc89970, c=<optimized out>, id=0, argv=0x7ffff10c40e0) at qdbusinterface.cpp:278
#10 0x00007f5e74a7f206 in QDBusConnectionPrivate::deliverCall (this=0xaea290, object=0xdf0540, msg=..., metaTypes=..., slotIdx=5) at qdbusintegrator.cpp:951
#11 0x00007f5e776fc9fe in QObject::event (this=0xdf0540, e=<optimized out>) at kernel/qobject.cpp:1194
#12 0x00007f5e76a856cc in QApplicationPrivate::notify_helper (this=this@entry=0x967710, receiver=receiver@entry=0xdf0540, e=e@entry=0xb58170) at kernel/qapplication.cpp:4567
*** Bug 331996 has been marked as a duplicate of this bug. *** *** Bug 332261 has been marked as a duplicate of this bug. *** *** Bug 332299 has been marked as a duplicate of this bug. *** *** Bug 333052 has been marked as a duplicate of this bug. *** *** Bug 333202 has been marked as a duplicate of this bug. *** *** Bug 333490 has been marked as a duplicate of this bug. *** Created attachment 86142 [details]
New crash information added by DrKonqi
kded4 (4.11.5) on KDE Platform 4.11.5 using Qt 4.8.5
- What I was doing when the application crashed:
stuff
i closed the laptop lid while personal settings application was open
-- Backtrace (Reduced):
#6 0x00007f65823c6aed in Solid::Backends::UPower::Battery::slotChanged (this=0x21d3e10) at /usr/src/debug/kdelibs-4.11.5/solid/solid/backends/upower/upowerbattery.cpp:139
[...]
#9 0x00007f6588cf07f2 in QDBusInterfacePrivate::metacall (this=0x1f70b90, c=<optimized out>, id=0, argv=0x7fffa75d4c90) at qdbusinterface.cpp:278
#10 0x00007f6588cddead in QDBusConnectionPrivate::deliverCall (this=0x1fe01c0, object=0x21d1720, msg=..., metaTypes=..., slotIdx=5) at qdbusintegrator.cpp:951
#11 0x00007f658bfad11e in QObject::event (this=0x21d1720, e=<optimized out>) at kernel/qobject.cpp:1203
#12 0x00007f658b1148ac in QApplicationPrivate::notify_helper (this=this@entry=0x1d84c00, receiver=receiver@entry=0x21d1720, e=e@entry=0x2223ef0) at kernel/qapplication.cpp:4562
*** Bug 333679 has been marked as a duplicate of this bug. *** Created attachment 86475 [details]
New crash information added by DrKonqi
kded4 (4.13.0) on KDE Platform 4.13.0 using Qt 4.8.5
- What I was doing when the application crashed:
krash was there at the resuming of the laptop
-- Backtrace (Reduced):
#6 0x00007f89dd0a632d in Solid::Backends::UPower::Battery::slotChanged (this=0x18fcc30) at /usr/src/debug/kdelibs-4.13.0/solid/solid/backends/upower/upowerbattery.cpp:149
[...]
#9 0x00007f89e12fa9f1 in QDBusInterfacePrivate::metacall (this=0x16ab160, c=<optimized out>, id=0, argv=0x7fffea0a1900) at qdbusinterface.cpp:278
#10 0x00007f89e12e5e2a in QDBusConnectionPrivate::deliverCall (this=0x14a6920, object=0x17d4180, msg=..., metaTypes=..., slotIdx=5) at qdbusintegrator.cpp:951
#11 0x00007f89e457692e in QObject::event (this=0x17d4180, e=<optimized out>) at kernel/qobject.cpp:1203
#12 0x00007f89e36ee92c in QApplicationPrivate::notify_helper(QObject*, QEvent*) () from /usr/lib64/libQtGui.so.4
*** Bug 334638 has been marked as a duplicate of this bug. *** *** Bug 334796 has been marked as a duplicate of this bug. *** *** Bug 335615 has been marked as a duplicate of this bug. *** In my case this is definitely associated with the "low batter" notification that I get when my battery drops below 20% while the laptop is sleeping. I can reproduce it, but of course each test takes a few hours. With the power disconnected, I suspended the laptop and woke it some hours later, with the battery draining by 4% to 11% during the interval. The outcomes were * Battery level above 20% when it was suspended and awakened: 6 trials, no crashes. * Battery level below 20% when it was suspended and awakened: 1 trial, no crash. * Battery level above 20% when it was suspended and below 20% when it woke: two more trials, two crashes. See bug 335615 for details of my case. *** Bug 336232 has been marked as a duplicate of this bug. *** *** Bug 336358 has been marked as a duplicate of this bug. *** Created attachment 87481 [details]
New crash information added by DrKonqi
kded4 (4.13.1) on KDE Platform 4.13.1 using Qt 4.8.6
It just happened again. Is there an ETA on resolving this issue?
- What I was doing when the application crashed:
Woke the computer from sleep without logging in directly. Went off to do other things, and then had to wake it once more (I think that was the 1st time this computer went back to sleep due to a login timeout). Logged in and started doing stuff immediately, and the crash followed.
-- Backtrace (Reduced):
#6 Solid::Backends::UPower::Battery::slotChanged (this=0x1606ef0) at ../../../solid/solid/backends/upower/upowerbattery.cpp:149
[...]
#9 0x00007f626fa92a82 in QDBusInterfacePrivate::metacall (this=0x167a5f0, c=<optimized out>, id=0, argv=0x7fff834d9680) at qdbusinterface.cpp:278
#10 0x00007f626fa801f6 in QDBusConnectionPrivate::deliverCall (this=0x1401990, object=0x1639aa0, msg=..., metaTypes=..., slotIdx=5) at qdbusintegrator.cpp:951
#11 0x00007f6272716c1e in QObject::event (this=0x1639aa0, e=<optimized out>) at kernel/qobject.cpp:1194
#12 0x00007f6271a97e2c in QApplicationPrivate::notify_helper (this=this@entry=0x11274e0, receiver=receiver@entry=0x1639aa0, e=e@entry=0x3170fa0) at kernel/qapplication.cpp:4567
Created attachment 87683 [details]
New crash information added by DrKonqi
kded4 (4.13.2) on KDE Platform 4.13.2 using Qt 4.8.6
- What I was doing when the application crashed:
Just woke the computer from suspension. Clearly this bug has survived the updates to 4.13.2 ....
-- Backtrace (Reduced):
#6 Solid::Backends::UPower::Battery::slotChanged (this=0x21cf5c0) at ../../../solid/solid/backends/upower/upowerbattery.cpp:139
[...]
#9 0x00007f4f8213fa82 in QDBusInterfacePrivate::metacall (this=0x21e15a0, c=<optimized out>, id=0, argv=0x7fff0e6be820) at qdbusinterface.cpp:278
#10 0x00007f4f8212d1f6 in QDBusConnectionPrivate::deliverCall (this=0x1ea0550, object=0x23c4030, msg=..., metaTypes=..., slotIdx=5) at qdbusintegrator.cpp:951
#11 0x00007f4f84dc3c1e in QObject::event (this=0x23c4030, e=<optimized out>) at kernel/qobject.cpp:1194
#12 0x00007f4f84144e2c in QApplicationPrivate::notify_helper (this=this@entry=0x1ab0500, receiver=receiver@entry=0x23c4030, e=e@entry=0x3fbd5a0) at kernel/qapplication.cpp:4567
Application: kded4 (4.13.3) KDE Platform Version: 4.13.3 Qt Version: 4.8.6 Operating System: Linux 3.15.6-200.fc20.x86_64 x86_64 Distribution: "Fedora release 20 (Heisenbug)" -- Information about the crash: - What I was doing when the application crashed: Woke the machine up from suspend (ram). Power cable was attached, and battery was >95 %. Battery widget stayed in the tray as before. Possible duplicates by query: bug 336358, bug 336232, bug 335615, bug 334796, bug 334638. *** Bug 338015 has been marked as a duplicate of this bug. *** *** Bug 339046 has been marked as a duplicate of this bug. *** I've experienced this crash many times, and as memory serves back as far as I first made a correlation, it always happens when the power adapter is changed after suspend, whether it went to sleep without power and then power was plugged in before resume (crashes less frequently, I think) or when the laptop has AC power when suspended, but power is unplugged before resume. This last time it happened as follows: put laptop to sleep overnight. Unplugged AC in one room. Plugged in AC in another room (different adapter, the *first* AC adapter is lower power enough that the bios complains about it at boot), when resumed, KDE daemon immediately crashed. I've experienced this crash many times, and as memory serves back as far as I first made a correlation, it always happens when the power adapter is changed after suspend, whether it went to sleep without power and then power was plugged in before resume (crashes less frequently, I think) or when the laptop has AC power when suspended, but power is unplugged before resume. This last time it happened as follows: put laptop to sleep overnight. Unplugged AC in one room. Plugged in AC in another room (different adapter, the *first* AC adapter is lower power enough that the bios complains about it at boot), when resumed, KDE daemon immediately crashed. Created attachment 88943 [details]
crash from ac adapter change after suspend
*** Bug 340697 has been marked as a duplicate of this bug. *** I have selected this bug as the "Bug of the Month". See https://community.kde.org/Gardening If you are able to fix it, you will receive a honorable mention in the next issue of my blog post "The Bug of the Month" on Planet KDE. Not all developers that would be able to fix it are subscribed to this bug. If you know someone, feel free to point him here. I haven't seen this happen in a while (on my netbook which is slow enough to make me not want to go bug hunting with it). Vontrapp may be onto something, because indeed I have been running less on battery, but have instead been moving the AC adapter to another socket, usually before waking the system. So most of the time there *is* a power change during suspend-to-RAM, but not a change in adapter, and also no change in the kind of power supply between the suspend and wake events. OTOH, I have begun to leave time to the system to "get the sleep out of its eyes" before beginning to use it, which more or less boils down to watching the WiFi indicator tell me what network it's connected to, and then (usually) change it to the more appropriate one. Does kded5 show the same issue? Apparently the signals can be bound to sth. w/ the potential to delete m_device. I'd suggest to const QString udi = m_device.data()->udi(); and use that string to emit the signals. This allows to fire all signals wile being robust against pointer invalidation. Created attachment 89817 [details]
Possible fix
I do not run into this, but if somebody does and can recompile kdelibs/solid, that somebody may want to try the attached patch.
(In reply to Thomas Lübking from comment #67) > Created attachment 89817 [details] > Possible fix > > I do not run into this, but if somebody does and can recompile > kdelibs/solid, that somebody may want to try the attached patch. Note/reminder to self: if I've got time & energy left for this after family visit tonight, update my kdelibs/git ppa :) Created attachment 89826 [details]
New crash information added by DrKonqi
kded4 (4.14.3) on KDE Platform 4.14.3 using Qt 4.8.6
- What I was doing when the application crashed:
Woke the computer from sleep.
Note that this time there has been no change in powersupply between the suspend and the wake, not even a power outage as far as I can tell. It is unclear to me why a battery-changed method would be invoked.
-- Backtrace (Reduced):
#6 Solid::Backends::UPower::Battery::slotChanged (this=0x2785590) at ../../../solid/solid/backends/upower/upowerbattery.cpp:149
[...]
#9 0x00007f328f4d54ba in QDBusInterfacePrivate::metacall (this=0x4ee9110, c=<optimized out>, id=0, argv=0x7fffdba8e1e0) at qdbusinterface.cpp:278
#10 0x00007f328f4c1b1e in QDBusConnectionPrivate::deliverCall (this=0x24034a0, object=0x2a2eb40, msg=..., metaTypes=..., slotIdx=5) at qdbusintegrator.cpp:951
#11 0x00007f329248135e in QObject::event (this=0x2a2eb40, e=<optimized out>) at kernel/qobject.cpp:1222
#12 0x00007f329171d0ac in QApplicationPrivate::notify_helper (this=this@entry=0x1f480f0, receiver=receiver@entry=0x2a2eb40, e=e@entry=0x26fc2d0) at kernel/qapplication.cpp:4570
Can someone who can still reproduce it state which distribution you are using? We could then ask them to prepare a test package in case you cannot compile the proposed patch. Without testing feedback, we will not be able to proceed further. (In reply to Christoph Feck from comment #70) > Can someone who can still reproduce it state which distribution you are > using? We could then ask them to prepare a test package in case you cannot > compile the proposed patch. Without testing feedback, we will not be able to > proceed further. I'm using Fedora 20 64-bit, and will be happy to use a test package. I've been running this patch almost since it was proposed, and have yet to have a crash. But I didn't have that many, so it's a bit early to cry victory. Patch is available for (K)Ubuntu 14.04 at ppa:rjvbertin/kde-git ; no need to ask ;) (In reply to Christoph Feck from comment #70) > Can someone who can still reproduce it state which distribution you are > using? We could then ask them to prepare a test package in case you cannot > compile the proposed patch. Without testing feedback, we will not be able to > proceed further. I definetely confirm that if you disable Power management in Service Manager - System Settings the problem disappear.. I'm using opensuse 13.1 and 13.2, I don't know how to apply a test package, so some instructions attached will be useful.. :-) :-) baseurl for openSUSE 13.2 users with Thomas' patch: http://download.opensuse.org/repositories/home:/sumski:/branches:/openSUSE:/13.2:/Update/openSUSE_13.2_Update/ Created attachment 90102 [details]
DrKonqi crash backtrace added manually
Seems I spoke too soon: kded4 just crashed on me ... but this was after 21 days of uptime of which a few days with 2 users logged with desktop sessions.
The backtrace is junk. -> Did you perhaps update packages (inc. kded)? Oh DAMN I swear I had a useful backtrace in front of me, but since the crash-after-wake came with absent WiFi (driver not loaded properly?) I couldn't send it through Konqi and thought that everything be saved when closing the dialog and then picking the proposed "save" option. Sigh. Created attachment 90219 [details]
New crash information added by DrKonqi
kded4 (4.14.3) on KDE Platform 4.14.3 using Qt 4.8.6
- What I was doing when the application crashed:
As usual, woke the computer from sleep. This time there ought to be a proper backtrace.
-- Backtrace (Reduced):
#7 0x00007f35169db890 in Solid::Backends::UPower::Battery::slotChanged (this=0x271a560) at ../../../solid/solid/backends/upower/upowerbattery.cpp:155
[...]
#10 0x00007f351d18fbe4 in QDBusInterfacePrivate::metacall (this=0x4f25590, c=<optimized out>, id=0, argv=0x7fff589e57d0) at qdbusinterface.cpp:278
#11 0x00007f351d180d2d in QDBusConnectionPrivate::deliverCall (this=0x2239c90, object=0x4f16210, msg=..., metaTypes=..., slotIdx=5) at qdbusintegrator.cpp:951
#12 0x00007f351ffd1d39 in QObject::event (this=0x4f16210, e=<optimized out>) at kernel/qobject.cpp:1222
#13 0x00007f351f37382c in QApplicationPrivate::notify_helper (this=0x1ee10f0, receiver=0x4f16210, e=0x266dae0) at kernel/qapplication.cpp:4570
René, if I understand you correctly, backtrace from comment #78 was made while the patch was applied, and confirms it does not fix the issue? Sadly, yes. Created attachment 90328 [details]
another possible fix
Ok, if it's not m_device, we probably loose the object itself due to such signal being invoked.
Interestingly, though, the crash moved down by one emit.
Attached is a patch that guards "this" - I wish everybody much fun with testing ;-)
Anyone interested and running (K)Ubuntu 14.04 can install the patched kdelibs packages from my PPA: ppa:rjvbertin/kde-git We still need feedback for the updated patch. See also comment #70. I have been running the latest patch since updating my PPA on Jan. 11th, on 2 machines that both see at least 1 suspend/wake cycle per day. I haven't had a crash - yet. I did have 2 occasions where kded4 exited without trace of a crash or error report. After the 1st time I have started the daemon from a terminal using `kdeinit4 kded4` but there was nothing interesting on that terminal after the 2nd exit. This bug is reported on libsolid which is the kdelibs4 version of the solid library. It is now in maintenance mode. If you think it should still be fixed in the KDE Frameworks 5 version of solid please move it to or report a bug on frameworks-solid. I have not been seeing kded4 crashes anymore, confirming what I already reported on Jan. 28th. As far as I'm concerned Thomas' patch can be committed as part of normal maintenance on KDE4. If kded5 crashes similarly I suggest to test that same patch as a first step, if it applies of course. The very same code altered by the patch also exists in KF5 - if there's a problem and if this problem is bypassed by the patch, it's very likely required for KF5 as well. I'm no longer seeing these crashes with kdelibs 4.14.6. I'm no longer seeing crashes either. What I am seeing is occasional exits of the kde daemon. They may be even more occasional than the crashes but they're certainly more subtle and even harder to debug. Is there a "silent and graceful" exit path in kded4 that is taken in certain conditions (possibly due to Thomas' fix), or are there conditions (e.g. during wake from sleep) under which a crash leads to such an exit rather than to a crash dialog? *** Bug 348107 has been marked as a duplicate of this bug. *** *** Bug 356091 has been marked as a duplicate of this bug. *** Created attachment 99216 [details]
Crash Report from error after awaking laptop from sleep
Hello! Sorry to be the bearer of bad news, but this project has been unmaintained for many years so I am closing this bug. Please try again with the latest version and submit a new bug to frameworks-solid if your issue persists. Thank you! For once "resolved" is probably appropriate because I haven't been seeing this crash since so long I completely forgot about the ticket. I wonder if this is also because all the crashes I did see and report were with Qt 4.8.6 or earlier, not the latest Qt 4.8.7 . (That doesn't mean there's a law against committing Thomas' patch of course, if it hasn't been already.) |