knetworkmanager crashed with SIGSEGV in Knm::Activatable::deviceUni()

Bug #453697 reported by Kai Blin
24
This bug affects 3 people
Affects Status Importance Assigned to Milestone
knetworkmanager
Fix Released
High
plasma-widget-networkmanagement (Ubuntu)
Incomplete
Low
Unassigned

Bug Description

Binary package hint: plasma-widget-networkmanagement

Crash looks similar to bug #408508, but that is reported as "fix released" on August 18th, and I just did a complete safe-upgrade of my Karmic system. Also, it happened while doing a restart of network-manager.

Opening a new bug so I can look at my stack trace again to compare to the bug #408508 stack trace.

ProblemType: Crash
Architecture: i386
Date: Sat Oct 17 06:23:13 2009
DistroRelease: Ubuntu 9.10
ExecutablePath: /usr/bin/knetworkmanager
Package: plasma-widget-networkmanagement 0.9~svn1029786-0ubuntu1
ProcCmdline: /usr/bin/knetworkmanager
ProcEnviron:
 PATH=(custom, user)
 LANG=en_US.UTF-8
 SHELL=/bin/bash
ProcVersionSignature: Ubuntu 2.6.31-14.47-generic
SegvAnalysis:
 Segfault happened at: 0xe3b80e <_ZNK3Knm11Activatable9deviceUniEv+14>: lock incl (%edx)
 PC (0x00e3b80e) ok
 source "(%edx)" (0x20796c70) not located in a known VMA region (needed readable region)!
SegvReason: reading unknown VMA
Signal: 11
SourcePackage: plasma-widget-networkmanagement
StacktraceTop:
 Knm::Activatable::deviceUni() const ()
 ?? () from /usr/lib/libknmservice.so.4
 ?? () from /usr/lib/libknmservice.so.4
 ?? () from /usr/lib/libknmservice.so.4
 SortedActivatableList::handleAdd(Knm::Activatable*) ()
Title: knetworkmanager crashed with SIGSEGV in Knm::Activatable::deviceUni()
Uname: Linux 2.6.31-14-generic i686
UserGroups: adm admin cdrom dialout lpadmin plugdev sambashare
XsessionErrors: (polkit-gnome-authentication-agent-1:3852): GLib-CRITICAL **: g_once_init_leave: assertion `initialization_value != 0' failed

Revision history for this message
Kai Blin (kai.blin) wrote :
Revision history for this message
Kai Blin (kai.blin) wrote :

Hm, the stack trace looks very similar to bug 408508. Are we sure that bug is fixed?

Revision history for this message
Apport retracing service (apport) wrote : Stacktrace.txt (retraced)

StacktraceTop:Knm::Activatable::deviceUni (this=0x87b8808)
compareDevices (first=0x87b8808, second=0x8768b30)
activatableLessThan (first=0x87b8808, second=0x8768b30)
qSortHelper<QList<Knm::Activatable*>::iterator, Knm::Activatable*, bool (*)(const Knm::Activatable*, const Knm::Activatable*)> (
SortedActivatableList::handleAdd (this=0x87ba320,

Revision history for this message
Apport retracing service (apport) wrote : ThreadStacktrace.txt (retraced)
Changed in plasma-widget-networkmanagement (Ubuntu):
importance: Undecided → Medium
tags: removed: need-i386-retrace
Revision history for this message
Kai Blin (kai.blin) wrote :

Thanks to the retracer, it's obvious that the "this" pointer in Knm::Activatable::deviceUni is not NULL in this stack trace. So it's a different issue after all.

Revision history for this message
Jonathan Thomas (echidnaman) wrote :

deviceUni is the only function that is the same between the two stacktraces. These are two separate crashes.

Changed in plasma-widget-networkmanagement (Ubuntu):
importance: Medium → Low
status: New → Confirmed
Revision history for this message
Jonathan Thomas (echidnaman) wrote :

Oh, I missed your comment where you determined the two stacktraces were different. Excuse me. :)

Changed in plasma-widget-networkmanagement (Ubuntu):
status: Confirmed → Triaged
visibility: private → public
1 comments hidden view all 120 comments
Revision history for this message
In , Jonathan Thomas (echidnaman) wrote :
Download full text (16.0 KiB)

Version: (using KDE 4.3.2)
OS: Linux
Installed from: Ubuntu Packages

Using KNetworkManager 0.9 svn 1029786, originally reported at https://launchpad.net/bugs/453697

The user reported a KNetworkManager crash that occurred when NetworkManager restarted during an upgrade. The crash had the following backtrace:

.
Thread 1 (process 3868):
#0 Knm::Activatable::deviceUni (this=0x87b8808)
    at /usr/include/qt4/QtCore/qatomic_i386.h:120
No locals.
#1 0x0037e0e1 in compareDevices (first=0x87b8808, second=0x8768b30)
    at /build/buildd/plasma-widget-networkmanagement-0.9~svn1029786/libs/service/sortedactivatablelist.cpp:263
 firstIface = (class Solid::Control::NetworkInterface *) 0x0
 secondIface = <value optimized out>
#2 0x0037e23c in activatableLessThan (first=0x87b8808, second=0x8768b30)
    at /build/buildd/plasma-widget-networkmanagement-0.9~svn1029786/libs/service/sortedactivatablelist.cpp:173
 i = <value optimized out>
#3 0x0037e502 in qSortHelper<QList<Knm::Activatable*>::iterator, Knm::Activatable*, bool (*)(const Knm::Activatable*, const Knm::Activatable*)> (
    start=<value optimized out>, end=DWARF-2 expression error: DW_OP_reg operations must be used either alone or in conjuction with DW_OP_piece.
)
    at /usr/include/qt4/QtCore/qalgorithms.h:384
 span = 12
#4 0x0037e6b7 in SortedActivatableList::handleAdd (this=0x87ba320,
    activatable=0x8768b30) at /usr/include/qt4/QtCore/qalgorithms.h:204
 d = (SortedActivatableListPrivate * const) 0x87ba2a8
#5 0x0036cd23 in ActivatableList::addActivatableInternal (this=0x8769b48,
    activatable=0x8768b30)
    at /build/buildd/plasma-widget-networkmanagement-0.9~svn1029786/libs/service/activatablelist.cpp:94
 observer = (class ActivatableObserver *) 0xbf945fec
 _container_ = {c = {{p = {static shared_null = {ref = {
            _q_value = 2338}, alloc = 0, begin = 0, end = 0, sharable = 1,
          array = {0x0}}, d = 0x87bb938}, d = 0x87bb938}}, brk = 0, i = {
    i = 0x87bb970}, e = {i = 0x87bb974}}
 d = <value optimized out>
#6 0x0036cdbc in ActivatableList::addActivatable (this=0x8769b48,
    activatable=0x20796c70)
    at /build/buildd/plasma-widget-networkmanagement-0.9~svn1029786/libs/service/activatablelist.cpp:82
No locals.
#7 0x00370db5 in WirelessNetworkInterfaceActivatableProvider::handleAdd (
    this=0x87e3fd0, addedConnection=0x87671c8)
    at /build/buildd/plasma-widget-networkmanagement-0.9~svn1029786/libs/service/wirelessnetworkinterfaceactivatableprovider.cpp:132
 ifaceConnection = (class Knm::WirelessInterfaceConnection *) 0x8768b30
 network = (class Solid::Control::WirelessNetwork *) 0x0
 knmNetwork = (class Knm::WirelessNetwork *) 0x0
 ourWicFound = false
 ourHiddenWicFound = false
 wirelessSetting = (class Knm::WirelessSetting *) 0x87ba870
 __PRETTY_FUNCTION__ = "virtual void WirelessNetworkInterfaceActivatableProvider::handleAdd(Knm::Connection*)"
#8 0x003710f9 in WirelessNetworkInterfaceActivatableProvider::networkAppeared
    (this=0x87e3fd0, ssid=@0xbf946428)
    at /build/buildd/plasma-widget-networkmanagement-0.9~svn1029786/libs/service/wirelessnetworkinterfaceactivatableprovider.cpp:176
 connection = (class Knm::Connecti...

Revision history for this message
In , Blin-q (blin-q) wrote :

Actually I was manually restarting network manager as it had failed to pick up my IPv6 router advertisement and thus didn't assign a global IPv6 address. So the crash actually happened during "normal operation" rather than during an upgrade.

Revision history for this message
In , Dario Andres (andresbajotierra) wrote :

*** Bug 212443 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Dario Andres (andresbajotierra) wrote :

*** Bug 212441 has been marked as a duplicate of this bug. ***

Revision history for this message
In , X-wstephenson (x-wstephenson) wrote :

I can't reproduce this on openSUSE; looks like I'm installing ubuntu this weekend then.

Revision history for this message
In , Dario Andres (andresbajotierra) wrote :

*** Bug 213237 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Jonathan Thomas (echidnaman) wrote :

*** Bug 213522 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Jonathan Thomas (echidnaman) wrote :

*** Bug 213775 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Dario Andres (andresbajotierra) wrote :

*** Bug 209092 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Dario Andres (andresbajotierra) wrote :

*** Bug 214428 has been marked as a duplicate of this bug. ***

Revision history for this message
In , X-wstephenson (x-wstephenson) wrote :

*** Bug 211586 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Dario Andres (andresbajotierra) wrote :

*** Bug 215781 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Dario Andres (andresbajotierra) wrote :

*** Bug 217089 has been marked as a duplicate of this bug. ***

Revision history for this message
In , ochach (ochach) wrote :

Created attachment 39160
knetworkmanager-20091218.kcrash

Revision history for this message
In , ochach (ochach) wrote :

Comment on attachment 39160
knetworkmanager-20091218.kcrash

restarting network-manager cause knetworkmanager crash every single time.

Revision history for this message
In , Dario Andres (andresbajotierra) wrote :

*** Bug 219396 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Dario Andres (andresbajotierra) wrote :

*** Bug 219689 has been marked as a duplicate of this bug. ***

Revision history for this message
In , X-wstephenson (x-wstephenson) wrote :

See also bko#215918

Revision history for this message
In , Dario Andres (andresbajotierra) wrote :

From bug 222257:
---
Some more information: The crash happened on my laptop using wifi and a wpa
secured network. If I disable wireless connections or use a wired connection
instead, then knetworkmanager does not crash. As soon as I enable wireless and
restart network-manager the crash appears.

Revision history for this message
In , Dario Andres (andresbajotierra) wrote :

*** Bug 222257 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Dario Andres (andresbajotierra) wrote :

*** Bug 222322 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Dario Andres (andresbajotierra) wrote :

Bug 224297 got some networkmanager-related config files.

Revision history for this message
In , Dario Andres (andresbajotierra) wrote :

*** Bug 224297 has been marked as a duplicate of this bug. ***

Changed in knetworkmanager:
status: Unknown → In Progress
Revision history for this message
In , Dario Andres (andresbajotierra) wrote :

*** Bug 224750 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Dario Andres (andresbajotierra) wrote :

*** Bug 226015 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Dario Andres (andresbajotierra) wrote :

*** Bug 226007 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Dario Andres (andresbajotierra) wrote :

*** Bug 227090 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Dario Andres (andresbajotierra) wrote :

*** Bug 231278 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Dario Andres (andresbajotierra) wrote :

*** Bug 232104 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Dario Andres (andresbajotierra) wrote :

*** Bug 231861 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Neoclust-kde (neoclust-kde) wrote :

*** Bug 235675 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Christoph-maxiom (christoph-maxiom) wrote :

*** Bug 242283 has been marked as a duplicate of this bug. ***

40 comments hidden view all 120 comments
Revision history for this message
In , Dario Andres (andresbajotierra) wrote :

*** Bug 258956 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Dario Andres (andresbajotierra) wrote :

*** Bug 259221 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Dario Andres (andresbajotierra) wrote :

*** Bug 260324 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Dario Andres (andresbajotierra) wrote :

*** Bug 260637 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Dario Andres (andresbajotierra) wrote :

*** Bug 263968 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Dario Andres (andresbajotierra) wrote :

*** Bug 261893 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Lamarque (lamarque) wrote :

This problem is caused by dangling pointers left in ActivableLists before NM starts again. Last November I commited a patch to workaround this problem by emptying the ActivatableList when NetworkManager goes down. It fixed the problem in my tests. Please upgrade to a recent networkmanagement (Plasma NM) revision.

Revision history for this message
In , Dario Andres (andresbajotierra) wrote :

*** Bug 264185 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Dario Andres (andresbajotierra) wrote :

*** Bug 264881 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Lamarque (lamarque) wrote :

*** Bug 258168 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Lamarque (lamarque) wrote :

*** Bug 263047 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Lamarque (lamarque) wrote :

*** Bug 256631 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Lamarque (lamarque) wrote :

*** Bug 265797 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Lamarque (lamarque) wrote :

Git commit 47519cb76d6285199ec2a6f0f8148490674c34a8 by Lamarque V. Souza.
Committed on 10/02/2011 at 08:34.
Pushed by lvsouza into branch 'master'.

Another try to solve the dangling pointers problem.

CCBUG: 210878

M +0 -21 libs/service/activatablelist.cpp
M +0 -2 libs/service/activatablelist.h
M +1 -0 solidcontrolfuture/wirelessnetworkinterfaceenvironment.cpp

http://commits.kde.org/networkmanagement/47519cb76d6285199ec2a6f0f8148490674c34a8

Revision history for this message
In , Lamarque (lamarque) wrote :

Git commit a8d87d386293681e3700275798984e8e3d602b79 by Lamarque V. Souza.
Committed on 10/02/2011 at 20:39.
Pushed by lvsouza into branch 'master'.

Prevents a crazy flux of several networkDisappeared signals for the same ssid
when several Solid::Control::WirelessNetworkInterfaceEnvironment
objects are deleted during wifi or NetworkManager turning off. The end
result is the same as using wirelessEnabledChanged(false) but now there
is no signal being emitted.

CCBUG: 210878

M +24 -0 libs/service/wirelessnetworkinterfaceactivatableprovider.cpp
M +0 -1 solidcontrolfuture/wirelessnetworkinterfaceenvironment.cpp

http://commits.kde.org/networkmanagement/a8d87d386293681e3700275798984e8e3d602b79

Revision history for this message
In , Frank78ac (frank78ac) wrote :

*** Bug 249226 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Frank78ac (frank78ac) wrote :

*** Bug 249270 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Frank78ac (frank78ac) wrote :

*** Bug 263487 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Frank78ac (frank78ac) wrote :

*** Bug 258469 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Lamarque (lamarque) wrote :

*** Bug 266079 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Lamarque (lamarque) wrote :

Please try the latest Plasma NM snapshot and report if this bug still happens. I would like to close this bug if it does not happen anymore.

Changed in knetworkmanager:
importance: Unknown → High
Revision history for this message
In , Lamarque (lamarque) wrote :

*** Bug 267500 has been marked as a duplicate of this bug. ***

Revision history for this message
In , grawcho (grawcho) wrote :

Created attachment 57837
New crash information added by DrKonqi

kded4 ($Id$) on KDE Platform 4.6.1 (4.6.1) using Qt 4.7.0

- What I was doing when the application crashed: kdm frozeup ... i restarted kdm and network manager did not restart. i restarted kdm manually ... and it still did not restart. i dialed via vwdial to report the bug.

-- Backtrace (Reduced):
#7 ref (this=0x9b18aa8) at /usr/include/qt4/QtCore/qatomic_i386.h:120
#8 QString (this=0x9b18aa8) at /usr/include/qt4/QtCore/qstring.h:728
#9 Knm::Activatable::deviceUni (this=0x9b18aa8) at ../../../libs/internals/activatable.cpp:51
#10 0x02829300 in compareDevices (first=0x9b1ec10, second=0x9b18aa8) at ../../../libs/service/sortedactivatablelist.cpp:263
#11 0x0282940c in activatableLessThan (first=0x9b1ec10, second=0x9b18aa8) at ../../../libs/service/sortedactivatablelist.cpp:172

Revision history for this message
In , Lamarque (lamarque) wrote :

(In reply to comment #94)
> Created an attachment (id=57837) [details]
> New crash information added by DrKonqi
>
> kded4 ($Id$) on KDE Platform 4.6.1 (4.6.1) using Qt 4.7.0
>
> - What I was doing when the application crashed: kdm frozeup ... i restarted
> kdm and network manager did not restart. i restarted kdm manually ... and it
> still did not restart. i dialed via vwdial to report the bug.

This is odd. If you restarted kdm all kde session should have been restarted as well, including kded. It is very unlikely this bug happens on an fresh starting kded. NetworkManager is not supposed to restart because any of its clients (Plasma NM, nm-appplet, whatever) have restarted. Are you sure what you restarted was kdm? You said kdm frozen, I have never seem kdm freezing. What were you doing before the freezing?

> -- Backtrace (Reduced):
> #7 ref (this=0x9b18aa8) at /usr/include/qt4/QtCore/qatomic_i386.h:120
> #8 QString (this=0x9b18aa8) at /usr/include/qt4/QtCore/qstring.h:728
> #9 Knm::Activatable::deviceUni (this=0x9b18aa8) at
> ../../../libs/internals/activatable.cpp:51
> #10 0x02829300 in compareDevices (first=0x9b1ec10, second=0x9b18aa8) at
> ../../../libs/service/sortedactivatablelist.cpp:263
> #11 0x0282940c in activatableLessThan (first=0x9b1ec10, second=0x9b18aa8) at
> ../../../libs/service/sortedactivatablelist.cpp:172

Revision history for this message
In , Lamarque (lamarque) wrote :

*** Bug 256900 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Lamarque (lamarque) wrote :

*** Bug 268795 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Lamarque (lamarque) wrote :

*** Bug 215918 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Lamarque (lamarque) wrote :

*** Bug 256552 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Lamarque (lamarque) wrote :

*** Bug 270705 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Lamarque (lamarque) wrote :

*** Bug 263827 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Matthew Flaschen (matthew-flaschen) wrote :

Created attachment 58897
New crash information added by DrKonqi

kded4 ($Id: kded.cpp 1156841 2010-07-29 19:59:05Z zander $) on KDE Platform 4.5.1 (KDE 4.5.1) using Qt 4.7.0

KNetworkManager widget crashed with a segfault when I restarted network-manager. I was not upgrading.

-- Backtrace (Reduced):
#6 ref (this=0x124a850) at /usr/include/qt4/QtCore/qatomic_x86_64.h:121
#7 QString (this=0x124a850) at /usr/include/qt4/QtCore/qstring.h:728
#8 Knm::Activatable::deviceUni (this=0x124a850) at ../../../libs/internals/activatable.cpp:51
#9 0x00007f709d3f6ddd in compareDevices (first=0x124a850, second=0xddfcc0) at ../../../libs/service/sortedactivatablelist.cpp:262
#10 0x00007f709d3f6f03 in activatableLessThan (first=0x7fff6be52930, second=0x124a850) at ../../../libs/service/sortedactivatablelist.cpp:172

Revision history for this message
In , Lamarque (lamarque) wrote :

*** Bug 271666 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Lamarque (lamarque) wrote :

*** Bug 249299 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Lamarque (lamarque) wrote :

*** Bug 273684 has been marked as a duplicate of this bug. ***

Revision history for this message
In , Lamarque (lamarque) wrote :

Can someone reproduce this problem with a recent Plasma NM snapshot? Please keep in mind still today some distributions ship a very old version of Plasma NM from last year. When testing please use a recent version, in the past three months we fixed tens of bugs in Plasma NM, so you must test with a recent version.

Revision history for this message
In , Brunofr (brunofr) wrote :

Not my case with a PNM 0.9.2.60git.1307285472 stack
kde 4.6.3 openSUSE_Factory

Revision history for this message
In , Alexandr Kara (alexandr-kara) wrote :

I think the issue is long gone, I haven't noticed it for maybe 6 months or a year. I have just verified in KDE 4.6.3 and found no problems.

Revision history for this message
In , Lamarque (lamarque) wrote :

Ok, closing this bug then.

Changed in knetworkmanager:
status: In Progress → Fix Released
Revision history for this message
In , Christoph-maxiom (christoph-maxiom) wrote :

*** Bug 281685 has been marked as a duplicate of this bug. ***

110 comments hidden view all 120 comments
Revision history for this message
Maarten Bezemer (veger) wrote :

Thank you for taking the time to report this bug and helping to make Ubuntu better. We are sorry that we do not always have the capacity to look at all reported bugs in a timely manner. There have been many changes in Ubuntu since that time you reported the bug and your problem may have been fixed with some of the updates. It would help us a lot if you could test it on a currently supported Ubuntu version. When you test it and it is still an issue, kindly upload the updated logs by running apport-collect 453697 and any other logs that are relevant for this particular issue.

Changed in plasma-widget-networkmanagement (Ubuntu):
status: Triaged → Incomplete
111 comments hidden view all 120 comments
Revision history for this message
In , Lamarque (lamarque) wrote :

*** Bug 325694 has been marked as a duplicate of this bug. ***

Displaying first 40 and last 40 comments. View all 120 comments or add a comment.
This report contains Public information  
Everyone can see this information.

Duplicates of this bug

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.