| Summary: | plasma crashes due to networkmanager | ||
|---|---|---|---|
| Product: | [Unmaintained] Network Management | Reporter: | kavol <kavol> |
| Component: | Plasma Widget | Assignee: | Sebastian Kügler <sebas> |
| Status: | RESOLVED FIXED | ||
| Severity: | crash | CC: | bruno, danlo8600, djidanetribal62, eddojh, esmb, grakkio, kavol, lamarque, montosh.bisht, mozinsur, pharrivel, rdieter, thilo, valir, wstephenson, wzab01 |
| Priority: | NOR | ||
| Version: | unspecified | ||
| Target Milestone: | --- | ||
| Platform: | Fedora RPMs | ||
| OS: | Linux | ||
| Latest Commit: | Version Fixed In: | 0.8.98 (0.9.0_rc3) | |
| Sentry Crash Report: | |||
| Attachments: |
New crash information added by DrKonqi
New crash information added by DrKonqi Test if pointer is valid before using it New crash information added by DrKonqi New crash information added by DrKonqi Valgrind log Defaults to InterfaceConnection instead of WirelessInterfaceConnection New crash information added by DrKonqi |
||
|
Description
kavol
2011-10-03 12:33:11 UTC
Created attachment 64241 [details]
New crash information added by DrKonqi
plasma-desktop (0.4) on KDE Platform 4.6.5 (4.6.5) using Qt 4.7.4
- What I was doing when the application crashed:
Unlike last time, the phone wasn't stuck - I just rebooted it while it was connected, which was enough to trigger the plasma crash.
A few lines from dmesg around the time of crash:
[18283.339695] usb 1-1.5.3: USB disconnect, device number 13
[18283.339777] cdc_ether 1-1.5.3:1.0: eth1: unregister 'cdc_ether' usb-0000:00:1a.0-1.5.3, CDC Ethernet Device
[18300.370293] usb 1-1.5.3: new full speed USB device number 14 using ehci_hcd
[18300.458683] usb 1-1.5.3: New USB device found, idVendor=0525, idProduct=a4a2
[18300.458689] usb 1-1.5.3: New USB device strings: Mfr=1, Product=2, SerialNumber=0
[18300.458694] usb 1-1.5.3: Product: RNDIS/Ethernet Gadget
[18300.458697] usb 1-1.5.3: Manufacturer: Linux 2.6.37.6 with s3c2410_udc
[18300.460126] cdc_subset: probe of 1-1.5.3:1.0 failed with error -22
[18300.460976] cdc_ether 1-1.5.3:1.0: eth1: register 'cdc_ether' at usb-0000:00:1a.0-1.5.3, CDC Ethernet Device, 00:1f:11:01:10:db
[18311.050170] eth1: no IPv6 routers present
-- Backtrace (Reduced):
#6 0x00007f74cb86ca42 in init (wobj=0x1866b08, this=0x1bfa1f0) at /usr/src/debug/networkmanagement-0.9/applet/wirelessstatus.cpp:65
#7 WirelessStatus::WirelessStatus (this=0x17c1f00, remote=0x1866af0) at /usr/src/debug/networkmanagement-0.9/applet/wirelessstatus.cpp:116
#8 0x00007f74cb86b985 in WirelessNetworkItem::WirelessNetworkItem (this=0x17e3170, remote=0x1866af0, parent=<optimized out>) at /usr/src/debug/networkmanagement-0.9/applet/wirelessnetworkitem.cpp:54
#9 0x00007f74cb85e60b in ActivatableListWidget::createItem (this=0xcd5140, activatable=0x1866af0, addIfAlreadyCached=<optimized out>) at /usr/src/debug/networkmanagement-0.9/applet/activatablelistwidget.cpp:178
#10 0x00007f74cb85e88d in ActivatableListWidget::activatableAdded (this=0xcd5140, added=0x1866af0) at /usr/src/debug/networkmanagement-0.9/applet/activatablelistwidget.cpp:275
I cannot reproduce this problem. Which Plasma NM version do you use? Which NM version do you use? This crash happens in the wireless code but you have not mentioned which wifi card you have nor if it was active when the crash happened. (In reply to comment #2) > I cannot reproduce this problem. unfortunately, it seems very irregular > Which Plasma NM version do you use? kde-plasma-networkmanagement-0.9-0.59.20110919git.nm09.fc15.x86_64 kde-plasma-networkmanagement-libs-0.9-0.59.20110919git.nm09.fc15.x86_64 > Which NM version do you use? now it is: NetworkManager-0.9.1.90-3.git20110927.fc15.x86_64 NetworkManager-glib-0.9.1.90-3.git20110927.fc15.x86_64 but it seems the upgrade is newer than the date of the crashes, I believe I had this before: NetworkManager-glib-0.9.0-1.fc15.x86_64 NetworkManager-0.9.0-1.fc15.x86_64 > This crash happens in the wireless code but you have not mentioned which wifi > card you have it detects like this: [ 61.824339] iwlagn 0000:03:00.0: Detected Intel(R) Centrino(R) Ultimate-N 6300 AGN, REV=0x74 > nor if it was active when the crash happened. if only I knew ... I got the machine docked, i.e. with ethernet cable plugged in, but I really do not know whether NM tried to activate wifi in parallel or not Created attachment 64458 [details]
New crash information added by DrKonqi
plasma-desktop (0.4) on KDE Platform 4.6.5 (4.6.5) using Qt 4.7.4
- What I was doing when the application crashed:
It just crashed again with the newer NetworkManager version (see above). Simply rebooting FreeRunner was enough to trigger the crash.
-- Backtrace (Reduced):
#6 0x00007f384d2b1a42 in init (wobj=0x26d3888, this=0x36b7be0) at /usr/src/debug/networkmanagement-0.9/applet/wirelessstatus.cpp:65
#7 WirelessStatus::WirelessStatus (this=0x33d4270, remote=0x26d3870) at /usr/src/debug/networkmanagement-0.9/applet/wirelessstatus.cpp:116
#8 0x00007f384d2b0985 in WirelessNetworkItem::WirelessNetworkItem (this=0x33e9710, remote=0x26d3870, parent=<optimized out>) at /usr/src/debug/networkmanagement-0.9/applet/wirelessnetworkitem.cpp:54
#9 0x00007f384d2a360b in ActivatableListWidget::createItem (this=0x2db3850, activatable=0x26d3870, addIfAlreadyCached=<optimized out>) at /usr/src/debug/networkmanagement-0.9/applet/activatablelistwidget.cpp:178
#10 0x00007f384d2a388d in ActivatableListWidget::activatableAdded (this=0x2db3850, added=0x26d3870) at /usr/src/debug/networkmanagement-0.9/applet/activatablelistwidget.cpp:275
(In reply to comment #4) > Created an attachment (id=64458) [details] > New crash information added by DrKonqi > > plasma-desktop (0.4) on KDE Platform 4.6.5 (4.6.5) using Qt 4.7.4 > > - What I was doing when the application crashed: > > It just crashed again with the newer NetworkManager version (see above). Simply > rebooting FreeRunner was enough to trigger the crash. Is FreeRunner a mobile phone? Please upgrade to a newer snapshot. Last month I made several changes in how activatables are initialized. *** Bug 284007 has been marked as a duplicate of this bug. *** Can you try upgrading your NM version? It seems it helps to solve this problem: https://bugzilla.redhat.com/show_bug.cgi?format=multiple&id=743891 (In reply to comment #5) > Is FreeRunner a mobile phone? well, they say so ... > Please upgrade to a newer snapshot. Last month I made several changes in how > activatables are initialized. there doesn't seem to be packaged anything newer than kde-plasma-networkmanagement-0.9-0.59.20110919git and I do not feel like messing with my install do you have any concrete suspicion or is that just a random try? (In reply to comment #7) > Can you try upgrading your NM version? It seems it helps to solve this problem: > https://bugzilla.redhat.com/show_bug.cgi?format=multiple&id=743891 comment #4 to RHBZ#743891 says "now on live image NetworkManager-0.9.1.90-3.git20110927.fc16 and no crash" I can hardly imagine how "upgrading" from NM 0.9.1.90-3.git20110927 (see comments #3 and #4 here) to NM 0.9.1.90-3.git20110927 could help ;-) as for upgrading to KDE 4.7.2, the same as above - F16 is not ready yet, or at least not in good enough shape for me to upgrade my system, I'm getting unresolvable package conflicts so I'll try to remember to report back once F16 with KDE 4.7.2 comes out (In reply to comment #8) > (In reply to comment #5) > > Is FreeRunner a mobile phone? > there doesn't seem to be packaged anything newer than > kde-plasma-networkmanagement-0.9-0.59.20110919git and I do not feel like > messing with my install > > do you have any concrete suspicion or is that just a random try? Sure it is a random try. There have been several changes in the part of the source code that deals with activatables like wirelessstatus.cpp. I also cannot reproduce this problem and all three people I know that have this problem uses Fedora packages. The only thing that would cause a crash like the one in your crashlog is a corrupted pointer. But I really do not know how a corrupted pointer could appear only in Plasma NM compiled by Fedora. I also do not understand what your phone has to do with this part of the source code, which deals with wifi access points not mobile broadband. > I can hardly imagine how "upgrading" from NM 0.9.1.90-3.git20110927 (see > comments #3 and #4 here) to NM 0.9.1.90-3.git20110927 could help ;-) Ok, so forget it. > as for upgrading to KDE 4.7.2, the same as above - F16 is not ready yet, or at > least not in good enough shape for me to upgrade my system, I'm getting > unresolvable package conflicts Problems in KDE SC rarely affects Plasma NM, upgrading to 4.7.2 is very unlikely to solve your problem. *** Bug 284284 has been marked as a duplicate of this bug. *** Please send me the output of the command: g++ -v For me the output is: $ g++ -v Using built-in specs. COLLECT_GCC=/usr/bin/g++ COLLECT_LTO_WRAPPER=/usr/libexec/gcc/x86_64-redhat-linux/4.6.1/lto-wrapper Target: x86_64-redhat-linux Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --with-bugurl=http://bugzilla.redhat.com/bugzilla --enable-bootstrap --enable-shared --enable-threads=posix --enable-checking=release --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-gnu-unique-object --enable-linker-build-id --enable-languages=c,c++,objc,obj-c++,java,fortran,ada,go,lto --enable-plugin --enable-java-awt=gtk --disable-dssi --with-java-home=/usr/lib/jvm/java-1.5.0-gcj-1.5.0.0/jre --enable-libgcj-multifile --enable-java-maintainer-mode --with-ecj-jar=/usr/share/java/eclipse-ecj.jar --disable-libjava-multilib --with-ppl --with-cloog --with-tune=generic --with-arch_32=i686 --build=x86_64-redhat-linux Thread model: posix gcc version 4.6.1 20110908 (Red Hat 4.6.1-9) (GCC) By the way I have the same problem (see duplicate) with a device which brings up a USB Gadget Ethernet (as FreeRunner does it seems). My wifi card is: 2:00.0 Network controller: Atheros Communications Inc. AR9285 Wireless Network Adapter (PCI-Express) (rev 01) Same NM version: NetworkManager-0.9.1.90-3.git20110927.fc15.x86_64 Created attachment 64678 [details]
Test if pointer is valid before using it
Well, g++ 4.6.1 is a bit experimental IMO. Maybe it is doing something wrong. Anyway, can someone try the attached patch and see if it helps to prevent the crash?
(In reply to comment #13) > Created an attachment (id=64678) [details] > Test if pointer is valid before using it > > Well, g++ 4.6.1 is a bit experimental IMO. Maybe it is doing something wrong. > Anyway, can someone try the attached patch and see if it helps to prevent the > crash? Thank you for the patch, I'll be abroad so I cannot test it now, please be patient! Created attachment 64809 [details]
New crash information added by DrKonqi
plasma-desktop (0.4) on KDE Platform 4.7.2 (4.7.2) using Qt 4.7.4
- What I was doing when the application crashed:
I connect my N9 by usb, and push the choice connect & synchronize. Immediately Plasma Desktop Crash
If I connect it as storage device it's ok, plasma doesn't crash ...
-- Backtrace (Reduced):
#6 0x00007f3f07ffc902 in init (wobj=0x2fd6108, this=0x2d20620) at /usr/src/debug/networkmanagement/applet/wirelessstatus.cpp:65
#7 WirelessStatus::WirelessStatus (this=0x2fd0360, remote=0x2fd60f0) at /usr/src/debug/networkmanagement/applet/wirelessstatus.cpp:116
#8 0x00007f3f07ffbf75 in WirelessNetworkItem::WirelessNetworkItem (this=0x2d84520, remote=0x2fd60f0, parent=<optimized out>) at /usr/src/debug/networkmanagement/applet/wirelessnetworkitem.cpp:54
#9 0x00007f3f07fededb in ActivatableListWidget::createItem (this=0x174e010, activatable=0x2fd60f0, addIfAlreadyCached=<optimized out>) at /usr/src/debug/networkmanagement/applet/activatablelistwidget.cpp:183
#10 0x00007f3f07fee15d in ActivatableListWidget::activatableAdded (this=0x174e010, added=0x2fd60f0) at /usr/src/debug/networkmanagement/applet/activatablelistwidget.cpp:280
Guys, please, when reporting a bug in Plasma NM please add the Plasma NM and NetworkManager version you are using and which distribution. By the crashlog I see you all are using a version of Plasma NM at least one month old. During this time the part of the source code when the crash happens was refactored. Can you please upgrade to the latest git version (nm09 branch) and see if the problem still happens? Okay I saw that Will is on CC, @Lamarque, sorry to use released software, most of end users act like that. If you are not happy with the result of drKonqui patches welcomed. drKonqui is not able to add individual kded's modules and plasmoids versions, it would be great if it could. I rely on users to tell me the Plasma NM version they are using. Of course, drKonqui has no knowledge about non-KDE software, such as NM, so the same applies. I guess this problem has something to do with the Bluedevil 1.2's bug that crashes kded very often (http://bugs.kde.org/284717). I fixed two bugs yesterday to prevent Plasma NM from crashing when kded crashes. Unfortunately, I read the bug entries after I had released Plasma NM 0.8.90, so I need you to test the latest nm09 branch to see if those two fixes also fix this bug. *** Bug 285555 has been marked as a duplicate of this bug. *** Git commit 49fb02e3894ef7e96bf2856f262d3b14dfda79fd by Lamarque V. Souza. Committed on 03/11/2011 at 20:33. Pushed by lvsouza into branch 'master'. Test if DBus interface object is valid before using it. CCBUG: 283241 M +8 -1 applet/wirelessstatus.cpp M +5 -0 libs/client/remotewirelessinterfaceconnection.cpp M +1 -0 libs/client/remotewirelessinterfaceconnection.h M +6 -0 libs/client/remotewirelessnetwork.cpp M +1 -0 libs/client/remotewirelessnetwork.h M +1 -0 libs/client/remotewirelessobject.h M +1 -1 plasma_nm_version.h http://commits.kde.org/networkmanagement/49fb02e3894ef7e96bf2856f262d3b14dfda79fd Git commit 1240c6f6f83d6a0502f659c7c49fe6d03ecb917f by Lamarque V. Souza. Committed on 03/11/2011 at 20:33. Pushed by lvsouza into branch 'nm09'. Test if DBus interface object is valid before using it. CCBUG: 283241 M +8 -1 applet/wirelessstatus.cpp M +5 -0 libs/client/remotewirelessinterfaceconnection.cpp M +1 -0 libs/client/remotewirelessinterfaceconnection.h M +6 -0 libs/client/remotewirelessnetwork.cpp M +1 -0 libs/client/remotewirelessnetwork.h M +1 -0 libs/client/remotewirelessobject.h http://commits.kde.org/networkmanagement/1240c6f6f83d6a0502f659c7c49fe6d03ecb917f Can someone test if the last commit fix the bug? I have never been able to reliably reproduce it. Created attachment 65306 [details]
New crash information added by DrKonqi
plasma-desktop (0.4) on KDE Platform 4.7.2 (4.7.2) using Qt 4.7.4
- What I was doing when the application crashed:
Woke up my machine from s2ram. This happened with version 0.9.0rc2. Should the fix mentioned above already be included there?
-- Backtrace (Reduced):
#6 0x00007f8ed2b7af55 in init (wobj=0x2ef5638, this=0x37317c0) at ../../applet/wirelessstatus.cpp:67
#7 WirelessStatus::WirelessStatus (this=0x2ef8100, remote=0x2ef5620) at ../../applet/wirelessstatus.cpp:123
#8 0x00007f8ed2b799c4 in WirelessNetworkItem::WirelessNetworkItem (this=0x1bdd300, remote=0x2ef5620, parent=<optimized out>) at ../../applet/wirelessnetworkitem.cpp:53
#9 0x00007f8ed2b6c3f5 in ActivatableListWidget::createItem (this=0x1910480, activatable=0x2ef5620, index=0) at ../../applet/activatablelistwidget.cpp:181
#10 0x00007f8ed2b6c63e in ActivatableListWidget::activatableAdded (this=0x1910480, added=0x2ef5620, index=0) at ../../applet/activatablelistwidget.cpp:269
Yes, it is included. Actually the crash happens in the test I added to check if the DBus interface is valid. One question. Do you all have pm-utils installed? I use pm-utils here and it signals NM to sleep on suspend to disk / ram. When NM is sleeping Plasma NM clears all activatables (the items in the plasmoid's connection list). I guess if NM is not put to sleep that may cause dangling pointers that could cause this crash. In my notebook after I trigger the suspend to ram shortcut the system tray icon changes to a red square with a white cross in the middle, which means Plasma NM has disabled networking because NM is not available anymore. If that does not happen in you notebook then that can be the problem. I guess I may found the problem. In bluedevil/src/actionplugins/networkpanu/helper/networkpanuhelper.cpp Bluedevil search only for the PANU service (00001115-0000-1000-8000-00805F9B34FB), but there is also the NAP service which uses UUID 00001116-0000-1000-8000-00805F9B34FB. In Plasma NM I only test for the NAP service, which is the one that worked with my friend's iPhone 3GS serveral months ago. You can try registering your tether connection using the networkmanagement_configshell command line tool: http://lamarque-lvs.blogspot.com/2011/04/plasma-nm-bluetooth-tethering.html if it works then the problem described in comment #27 is the source of the problem. On Sun, Nov 6, 2011 at 19:08, Lamarque V. Souza <lamarque@kde.org> wrote: > --- Comment #26 from Lamarque V. Souza <lamarque kde org> 2011-11-06 18:08:55 --- > One question. Do you all have pm-utils installed? Yes. > I use pm-utils here and it > signals NM to sleep on suspend to disk / ram. When NM is sleeping Plasma NM > clears all activatables (the items in the plasmoid's connection list). I guess > if NM is not put to sleep that may cause dangling pointers that could cause > this crash. > > In my notebook after I trigger the suspend to ram shortcut the system tray icon > changes to a red square with a white cross in the middle, which means Plasma NM > has disabled networking because NM is not available anymore. If that does not > happen in you notebook then that can be the problem. Frankly, the suspend happens too fast, so I cannot tell whether the icon changes. The crash happens always, so if you would like me to do some debugging, let me know. My NM connections are system connections if that makes a difference. BTW, I do not use Bluetooth tethering, but I do have a WWAN (UMTS) modem in my laptop. Sorry about the bluetooth tethering messages, they should have gone to another bug I am trying to solve. System connection should not make any difference, but you can try changing them to user connections just to make sure. Also try running plasma-desktop through valgrind: killall plasma-desktop valgrind -v --leak-check=full --log-file=plasma-desktop.log --malloc-fill=0xaa --free-fill=0xbb plasma-desktop You must be patient when using valgrind, it will make plasma-desktop start up extremelly slow (almost a minute). Compress the plasma-desktop.log file and attach it to the this bug entry. Created attachment 65318 [details]
Valgrind log
When I change my WiFi connection to a user connection, the NM Widget even sporadically crashes during login (but also during resume). Please find attached the requested valgrind log. I guess that having a connection being established is sufficient to trigger the issue.
Strange, according to the log Plasma NM is allocating memory for a VPN connection but doing a cast on a wireless connection, that should not happen. There are othe memory errors in the log that do not seem to be related to Plasma NM. I will try to investigate further. On Mon, Nov 7, 2011 at 00:51, Lamarque V. Souza <lamarque@kde.org> wrote: > --- Comment #32 from Lamarque V. Souza <lamarque kde org> 2011-11-06 23:51:12 --- > Strange, according to the log Plasma NM is allocating memory for a VPN > connection but doing a cast on a wireless connection, that should not happen. That's plausible. I do have an OpenVPN VPN connection defined, which is set to "Connect Automatically", but typically is not connected. Created attachment 65322 [details] Defaults to InterfaceConnection instead of WirelessInterfaceConnection (In reply to comment #33) > On Mon, Nov 7, 2011 at 00:51, Lamarque V. Souza <lamarque@kde.org> wrote: > > --- Comment #32 from Lamarque V. Souza <lamarque kde org> 2011-11-06 23:51:12 --- > > Strange, according to the log Plasma NM is allocating memory for a VPN > > connection but doing a cast on a wireless connection, that should not happen. > > That's plausible. I do have an OpenVPN VPN connection defined, which > is set to "Connect Automatically", but typically is not connected. The problem is that the cast only works for wireless connections and that is causing the crash. I do not understand why it is trying to cast a wireless connection since the connection that was just added is a VPN one. It is like it has changed the connection type in the middle of process of adding a connection. I have a guess for what is happening, can you test the attached patch for me? Git commit 0e87ab1a567aca695d655df8e423b5f9de35b2e3 by Lamarque V. Souza. Committed on 06/11/2011 at 22:54. Pushed by lvsouza into branch 'nm09'. Revert "Test if DBus interface object is valid before using it." 1240c6f6f83d6a0502f659c7c49fe6d03ecb917f does not prevent the crash it is was supposed to fix. This reverts commit 1240c6f6f83d6a0502f659c7c49fe6d03ecb917f. CCBUG: 283241 M +1 -8 applet/wirelessstatus.cpp M +0 -5 libs/client/remotewirelessinterfaceconnection.cpp M +0 -1 libs/client/remotewirelessinterfaceconnection.h M +0 -6 libs/client/remotewirelessnetwork.cpp M +0 -1 libs/client/remotewirelessnetwork.h M +0 -1 libs/client/remotewirelessobject.h M +1 -1 plasma_nm_version.h http://commits.kde.org/networkmanagement/0e87ab1a567aca695d655df8e423b5f9de35b2e3 Git commit d1d481d82d350083cedd3b040e5121a23247cc49 by Lamarque V. Souza. Committed on 06/11/2011 at 19:15. Pushed by lvsouza into branch 'master'. Revert "Test if DBus interface object is valid before using it." 49fb02e3894ef7e96bf2856f262d3b14dfda79fd does not prevent the crash it is was supposed to fix. This reverts commit 49fb02e3894ef7e96bf2856f262d3b14dfda79fd. CCBUG: 283241 M +1 -8 applet/wirelessstatus.cpp M +0 -5 libs/client/remotewirelessinterfaceconnection.cpp M +0 -1 libs/client/remotewirelessinterfaceconnection.h M +0 -6 libs/client/remotewirelessnetwork.cpp M +0 -1 libs/client/remotewirelessnetwork.h M +0 -1 libs/client/remotewirelessobject.h M +1 -1 plasma_nm_version.h http://commits.kde.org/networkmanagement/d1d481d82d350083cedd3b040e5121a23247cc49 *** Bug 286032 has been marked as a duplicate of this bug. *** On Mon, Nov 7, 2011 at 01:37, Lamarque V. Souza <lamarque@kde.org> wrote: > --- Comment #34 from Lamarque V. Souza <lamarque kde org> 2011-11-07 00:37:27 --- > Created an attachment (id=65322) > --> (http://bugs.kde.org/attachment.cgi?id=65322) > Defaults to InterfaceConnection instead of WirelessInterfaceConnection > [...] > I have a guess for what is happening, can you test the attached patch for me? Sure. The patch works just great, no more crashes on resume. Thanks, Thilo Git commit ca33e27e81395b4ce2f78b728ea3e3946bbb3f07 by Lamarque V. Souza. Committed on 08/11/2011 at 00:23. Pushed by lvsouza into branch 'master'. Make two calls synchronous and also make RemoteActivatable::activatableType() return a sane default in case of errors. activatableType() was returning Knm::Activatable::WirelessInterfaceConnection in case of error or when the assynchronous DBus call returned before the real value was available, which led to a crash in applet/wirelessstatus.cpp when the real activatable type was not WirelessInterfaceConnection ot WirelessNetwork. BUG: 283241 FIXED-IN: 0.9.0 M +10 -3 libs/client/remoteactivatable.cpp http://commits.kde.org/networkmanagement/ca33e27e81395b4ce2f78b728ea3e3946bbb3f07 Git commit 5d6050bf3e70930bfb80321f710613142c2db95a by Lamarque V. Souza. Committed on 08/11/2011 at 00:23. Pushed by lvsouza into branch 'nm09'. Make two calls synchronous and also make RemoteActivatable::activatableType() return a sane default in case of errors. activatableType() was returning Knm::Activatable::WirelessInterfaceConnection in case of error or when the assynchronous DBus call returned before the real value was available, which led to a crash in applet/wirelessstatus.cpp when the real activatable type was not WirelessInterfaceConnection ot WirelessNetwork. BUG: 283241 FIXED-IN: 0.9.0 (cherry picked from commit ca33e27e81395b4ce2f78b728ea3e3946bbb3f07) M +10 -3 libs/client/remoteactivatable.cpp http://commits.kde.org/networkmanagement/5d6050bf3e70930bfb80321f710613142c2db95a *** Bug 286511 has been marked as a duplicate of this bug. *** *** Bug 287089 has been marked as a duplicate of this bug. *** *** Bug 287357 has been marked as a duplicate of this bug. *** (In reply to comment #40) > Git commit 5d6050bf3e70930bfb80321f710613142c2db95a by Lamarque V. Souza. > Committed on 08/11/2011 at 00:23. > Pushed by lvsouza into branch 'nm09'. if this is included in kde-plasma-networkmanagement-libs-0.9-0.65.rc2.nm09.fc16.x86_64 kde-plasma-networkmanagement-0.9-0.65.rc2.nm09.fc16.x86_64 then it seems to fix the issue for me, thanks! (In reply to comment #44) > (In reply to comment #40) > > Git commit 5d6050bf3e70930bfb80321f710613142c2db95a by Lamarque V. Souza. > > Committed on 08/11/2011 at 00:23. > > Pushed by lvsouza into branch 'nm09'. > > if this is included in > > kde-plasma-networkmanagement-libs-0.9-0.65.rc2.nm09.fc16.x86_64 > kde-plasma-networkmanagement-0.9-0.65.rc2.nm09.fc16.x86_64 According to http://rpmfind.net//linux/RPM/fedora/updates/16/x86_64/kde-plasma-networkmanagement-0.9-0.65.rc2.nm09.fc16.x86_64.html that package was created four days before the commit, so it is not included. The problem acts like a race condition where one of the racers wins almost all the time, so it is not always reproducible. *** Bug 287825 has been marked as a duplicate of this bug. *** *** Bug 288117 has been marked as a duplicate of this bug. *** fedora 16 users, mind testing out: kde-plasma-networkmanagement-0.9-0.67.20111203git.nm09.fc16 http://koji.fedoraproject.org/koji/buildinfo?buildID=277228 (I'll queue it for updates-testing shortly) (In reply to comment #45) > According to > http://rpmfind.net//linux/RPM/fedora/updates/16/x86_64/kde-plasma-networkmanagement-0.9-0.65.rc2.nm09.fc16.x86_64.html > that package was created four days before the commit, so it is not included. > > The problem acts like a race condition where one of the racers wins almost all > the time, so it is not always reproducible. yep, I just got another crash ... seems I was just lucky not seeing it twice in a row :-/ going to try the new package ... (In reply to comment #48) > fedora 16 users, mind testing out: > kde-plasma-networkmanagement-0.9-0.67.20111203git.nm09.fc16 > http://koji.fedoraproject.org/koji/buildinfo?buildID=277228 thanks; I rebooted the phone five times and no crashes so far ... > (I'll queue it for updates-testing shortly) I still do not see it in bodhi ...? Created attachment 66477 [details]
New crash information added by DrKonqi
plasma-desktop (0.4) on KDE Platform 4.7.2 (4.7.2) "release 5" using Qt 4.7.4
- What I was doing when the application crashed:
I was restarting the wifi router while the connection was active. During router restart, I inserted the LAN cable.
-- Backtrace (Reduced):
#6 0xaaa71e3b in init (wobj=0x8c66a34, this=0x8bb15b0) at /usr/src/debug/networkmanagement/applet/wirelessstatus.cpp:65
#7 WirelessStatus::WirelessStatus (this=0x90ff518, remote=0x8c66a28) at /usr/src/debug/networkmanagement/applet/wirelessstatus.cpp:116
#8 0xaaa7050b in WirelessNetworkItem::WirelessNetworkItem (this=0x9294bf0, remote=0x8c66a28, parent=0x8b46140) at /usr/src/debug/networkmanagement/applet/wirelessnetworkitem.cpp:53
#9 0xaaa61518 in ActivatableListWidget::createItem (this=0x8b3f658, activatable=0x8c66a28, index=0) at /usr/src/debug/networkmanagement/applet/activatablelistwidget.cpp:181
#10 0xaaa617bc in ActivatableListWidget::activatableAdded (this=0x8b3f658, added=0x8c66a28, index=0) at /usr/src/debug/networkmanagement/applet/activatablelistwidget.cpp:269
*** Bug 290031 has been marked as a duplicate of this bug. *** *** Bug 291005 has been marked as a duplicate of this bug. *** *** Bug 291084 has been marked as a duplicate of this bug. *** *** Bug 292884 has been marked as a duplicate of this bug. *** |