aMule Forum
English => aMule crashes => Topic started by: Brezhonneg on February 29, 2008, 10:11:04 AM
-
It does crash every now and then. Installed from the db found on this forum on a Kubuntu 64bit system. I was not in front of the computer when this happened...
----------------------------=| BACKTRACE FOLLOWS: |=----------------------------
Current version is: aMule SVN using wxGTK2 v2.8.4 (Snapshot: Wed Jan 23 07:01:57 CET 2008)
Running on: Linux 2.6.22-14-generic x86_64
[2] std::basic_string<char, std::char_traits<char>, std::allocator<char> >::~basic_string() in amule [0x454e0f]
[3] wxFatalSignalHandler in /usr/lib/libwx_baseu-2.8.so.0[0x2ae3b435864c]
[4] ?? in /lib/libpthread.so.0 [0x2ae3b2e03100]
[5] wxMenuItemList::~wxMenuItemList() in amule [0x682442]
[6] wxDataObjectSimple::~wxDataObjectSimple() in amule [0x61b2f5]
[7] wxDataObjectSimple::~wxDataObjectSimple() in amule [0x61b751]
[8] wxDataObjectSimple::~wxDataObjectSimple() in amule [0x61bc26]
[9] wxEvtHandler::ProcessEventIfMatches(wxEventTableEntryBase const&, wxEvtHandler*, wxEvent&) in /usr/lib/libwx_baseu-2.8.so.0[0x2ae3b435466f]
[10] wxEventHashTable::HandleEvent(wxEvent&, wxEvtHandler*) in /usr/lib/libwx_baseu-2.8.so.0[0x2ae3b435480f]
[11] wxEvtHandler::ProcessEvent(wxEvent&) in /usr/lib/libwx_baseu-2.8.so.0[0x2ae3b4354959]
[12] wxTimerBase::Notify() in /usr/lib/libwx_gtk2u_core-2.8.so.0[0x2ae3b3cdbd76]
[13] ?? in /usr/lib/libwx_gtk2u_core-2.8.so.0 [0x2ae3b3bd2a14]
[14] ?? in /usr/lib/libglib-2.0.so.0 [0x2ae3b680d70b]
[15] g_main_context_dispatch in /usr/lib/libglib-2.0.so.0[0x2ae3b680cfd3]
[16] ?? in /usr/lib/libglib-2.0.so.0 [0x2ae3b68102dd]
[17] g_main_loop_run in /usr/lib/libglib-2.0.so.0[0x2ae3b68105ea]
[18] gtk_main in /usr/lib/libgtk-x11-2.0.so.0[0x2ae3b6e0c883]
[19] wxEventLoop::Run() in /usr/lib/libwx_gtk2u_core-2.8.so.0[0x2ae3b3bca50d]
[20] wxAppBase::MainLoop() in /usr/lib/libwx_gtk2u_core-2.8.so.0[0x2ae3b3c5bc8b]
[21] wxEntry(int&, wchar_t**) in /usr/lib/libwx_baseu-2.8.so.0[0x2ae3b42f380c]
[22] std::ostream::operator<<(bool) in amule [0x612ce6]
[23] __libc_start_main in /lib/libc.so.6[0x2ae3b4f99b44]
[24] wxAppBase::MainLoop() in amule[0x44f1a9]
--------------------------------------------------------------------------------
Abandon (core dumped)
-
Wahoo... I restarted and got a crash again in the time it took for me to post the previous report. Must be the same but heck, here it is:
----------------------------=| BACKTRACE FOLLOWS: |=----------------------------
Current version is: aMule SVN using wxGTK2 v2.8.4 (Snapshot: Wed Jan 23 07:01:57 CET 2008)
Running on: Linux 2.6.22-14-generic x86_64
[2] std::basic_string<char, std::char_traits<char>, std::allocator<char> >::~basic_string() in amule [0x454e0f]
[3] wxFatalSignalHandler in /usr/lib/libwx_baseu-2.8.so.0[0x2b787536164c]
[4] ?? in /lib/libpthread.so.0 [0x2b7873e0c100]
[5] wxMenuItemList::~wxMenuItemList() in amule [0x682442]
[6] wxDataObjectSimple::~wxDataObjectSimple() in amule [0x61b2f5]
[7] wxDataObjectSimple::~wxDataObjectSimple() in amule [0x61b751]
[8] wxDataObjectSimple::~wxDataObjectSimple() in amule [0x61bc26]
[9] wxEvtHandler::ProcessEventIfMatches(wxEventTableEntryBase const&, wxEvtHandler*, wxEvent&) in /usr/lib/libwx_baseu-2.8.so.0[0x2b787535d66f]
[10] wxEventHashTable::HandleEvent(wxEvent&, wxEvtHandler*) in /usr/lib/libwx_baseu-2.8.so.0[0x2b787535d80f]
[11] wxEvtHandler::ProcessEvent(wxEvent&) in /usr/lib/libwx_baseu-2.8.so.0[0x2b787535d959]
[12] wxTimerBase::Notify() in /usr/lib/libwx_gtk2u_core-2.8.so.0[0x2b7874ce4d76]
[13] ?? in /usr/lib/libwx_gtk2u_core-2.8.so.0 [0x2b7874bdba14]
[14] ?? in /usr/lib/libglib-2.0.so.0 [0x2b787781670b]
[15] g_main_context_dispatch in /usr/lib/libglib-2.0.so.0[0x2b7877815fd3]
[16] ?? in /usr/lib/libglib-2.0.so.0 [0x2b78778192dd]
[17] g_main_loop_run in /usr/lib/libglib-2.0.so.0[0x2b78778195ea]
[18] gtk_main in /usr/lib/libgtk-x11-2.0.so.0[0x2b7877e15883]
[19] wxEventLoop::Run() in /usr/lib/libwx_gtk2u_core-2.8.so.0[0x2b7874bd350d]
[20] wxAppBase::MainLoop() in /usr/lib/libwx_gtk2u_core-2.8.so.0[0x2b7874c64c8b]
[21] wxEntry(int&, wchar_t**) in /usr/lib/libwx_baseu-2.8.so.0[0x2b78752fc80c]
[22] std::ostream::operator<<(bool) in amule [0x612ce6]
[23] __libc_start_main in /lib/libc.so.6[0x2b7875fa2b44]
[24] wxAppBase::MainLoop() in amule[0x44f1a9]
--------------------------------------------------------------------------------
Abandon (core dumped)
It usually takes a lot longer between two crashes.
-
I checked my dmesg, and here is what I found:
UDP: bad checksum. From xxx.xxx.xxx.xxx:xxxxx to yyy.yyy.yyy.yyy:yyyyy ulen 44
I removed the IPs. The second one was my local IP (local network).
I don't know the timing for sure, so this may or may not be related with the crashes...
EDIT: Note that I only got one message like this, while I got two crashes... If this error message has to be related with any of the two crashes, it must be with the first one because the timing does not match the second one.
-
I have been getting very frequent crashes always with the same trace since yesterday. What seems to have triggered this string was the addition of two popular files (with many many sources) in my download list, while I only had like a dozen of less popular files (10-20 sources top) before. I stopped one of the two popular files, I will see if that decreases the crash occurence at least. aMule is virtually unusable with that many crashes.
As far as the UPD checkum thingy in the kernel log goes, there is definitely no clear realtion with the crash, or at least it is not direct.
-
Sadely, my test with stopping one of the two popular files was not very conclusive. Crashes kept happening seemingly randomly, then stopped. Since then, the first popular file download was completed, I removed it from the list and resumed the second file, still no crash in 48h. So, I am thinking that this crash is triggered by some random event, which occurence is more frequent when you connect to more sources. Perhaps one of those stupid automatic spam messages we receive from time to time from some sources? Well, it could be anything I guess, I am sorry this report is not very helpful. Let me know if you can think of any more info/test I could provide/carry out.
Cheers
-
Hi Brezhonneg,
You could try to compile aMule with debug information so that you are able to provide a more accurate backtrace. See this link for clues:
http://www.amule.org/wiki/index.php/Backtraces
Cheers!
-
I have very similar crashes, while amule is only uploading. This occurs with the amule cvs snapshots of the last 2-3 weeks. That's bad....
I'll follow the guidelines to produce more useful stack traces. For now I have this. I'm on amd64 architecture, but I can reproduce sistematically the crash also on a 32 bit chroot.
The distribution used is debian Etch with vollkstener packages compiled with pbuilder for amd64. I've also backported the 20080309 snapshot from debian experimental, but no difference.
I have also the core file, where I can send it?
I noticed that amule process size grows with the time (as usually :o).
amuled with the same configuration doesn't crashes, but starts to use the 100% of the cpu, and that's is even worse.
----------------------------------------------------------------------------------------------------
Initialising aMule
Checking if there is an instance already running...
No other instances are running.
ListenSocket: Ok.
Loading temp files from /home/emi/.aMule/Temp.
HTTP download thread started
All PartFiles Loaded.
Host: peerates.net:80
URL: http://peerates.net/servers.php
Response: 200 (Error: 0)
Download size: 1525
HTTP download thread ended
HTTP download thread started
Host: www.gruk.org:80
URL: http://www.gruk.org/server.met
Response: 200 (Error: 0)
Download size: 3521
HTTP download thread ended
--------------------------------------------------------------------------------
A fatal error has occurred and aMule has crashed.
Please assist us in fixing this problem by posting the backtrace below in our
'aMule Crashes' forum and include as much information as possible regarding the
circumstances of this crash. The forum is located here:
http://forum.amule.org/index.php?board=67.0
If possible, please try to generate a real backtrace of this crash:
http://www.amule.org/wiki/index.php/Backtraces
----------------------------=| BACKTRACE FOLLOWS: |=----------------------------
Current version is: aMule SVN using wxGTK2 v2.8.7 (Snapshot: Sun Mar 9 07:01:55 CET 2008)
Running on: Linux 2.6.18.8-xen x86_64
[2] wxThreadHelperThread::~wxThreadHelperThread() in amule [0x444145]
[3] wxFatalSignalHandler in /usr/lib/libwx_baseu-2.8.so.0[0x3598afa97c]
[4] ?? in /lib/libpthread.so.0 [0x359400c410]
[5] wxMenuItemList::~wxMenuItemList() in amule [0x570a38]
[6] wxDataObjectSimple::~wxDataObjectSimple() in amule [0x515386]
[7] wxDataObjectSimple::~wxDataObjectSimple() in amule [0x515ad2]
[8] wxEvtHandler::ProcessEventIfMatches(wxEventTableEntryBase const&, wxEvtHandler*, wxEvent&) in /usr/lib/libwx_baseu-2.8.so.0[0x3598af67df]
[9] wxEventHashTable::HandleEvent(wxEvent&, wxEvtHandler*) in /usr/lib/libwx_baseu-2.8.so.0[0x3598af697e]
[10] wxEvtHandler::ProcessEvent(wxEvent&) in /usr/lib/libwx_baseu-2.8.so.0[0x3598af6ac9]
[11] wxTimerBase::Notify() in /usr/lib/libwx_gtk2u_core-2.8.so.0[0x36094f27a6]
[12] ?? in /usr/lib/libwx_gtk2u_core-2.8.so.0 [0x36093e96b4]
[13] ?? in /usr/lib/libglib-2.0.so.0 [0x360782ce1b]
[14] g_main_context_dispatch in /usr/lib/libglib-2.0.so.0[0x360782c913]
[15] ?? in /usr/lib/libglib-2.0.so.0 [0x360782f75d]
[16] g_main_loop_run in /usr/lib/libglib-2.0.so.0[0x360782fa46]
[17] gtk_main in /usr/lib/libgtk-x11-2.0.so.0[0x3608b2b6b2]
[18] wxEventLoop::Run() in /usr/lib/libwx_gtk2u_core-2.8.so.0[0x36093e0e6d]
[19] wxAppBase::MainLoop() in /usr/lib/libwx_gtk2u_core-2.8.so.0[0x36094733bb]
[20] wxEntry(int&, wchar_t**) in /usr/lib/libwx_baseu-2.8.so.0[0x3598a95cbc]
[21] std::basic_string<char, std::char_traits<char>, std::allocator<char> > std::operator+<char, std::char_traits<char>, std::allocator<char> >(std::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) in amule [0x510882]
[22] __libc_start_main in /lib/libc.so.6[0x3593a1c4ca]
[23] __gxx_personality_v0 in amule[0x44339a]
-
Please use a more recent versión (lastest 2008-03-03):
http://forum.amule.org/index.php?topic=13700.0
-
Hey, I was using:
Current version is: aMule SVN using wxGTK2 v2.8.7 (Snapshot: Sun Mar 9 07:01:55 CET 2008)
that is newer than 2008-03-03...
anyway. I compiled it with debug informations, installed all debug symbols for wx and glib, and ran amule from gdb. I crashed after one day, under the same circumstances (only uploading, no downloads...).
Attached there are the backtraces. I have also the core dump, but it's big so tell me where I can upload it, if you need it.
Feel free to move this thread on the backtraces section of the forum....
Thanks
-
I forgot, as it seems to be a problem related to the tray icon, as the full backtrace says..... I've started amule under gdb and disabled the tray icon.
Let's wait until it crashes or perhaps let's see what happens.
-
Aren't the first backtrace and the last one differents? I mean, the first one says nothing about MuleTrayIcon, but the last one does.
Anyway, the MuleTrayIcon crash was previously reported here:
http://forum.amule.org/index.php?topic=13914.0
http://forum.amule.org/index.php?topic=14141.0
AFAIK, this crash is triggered when you get a message with geo-ip enabled (ie. having compiled with --enable-geoip). This is probably your random crash which is more likely to happen when you connect to more sources. The way to avoid the crash is just filtering all messages (or compiling without geo-ip).
In the other hand: it's a crash very easy to reproduce with two instances of aMule, could anyone confirm my theory??
Hope it helps!! ;)
-
so let's wait if the next crash will be of the first type or the second.... :o
-
so let's wait if the next crash will be of the first type or the second.... :o
Oops, I forgot to say that disabling TrayIcon had no effect: aMule keeps crashing on arrival of a message.
I can't imagine how these 3 things are related: a crash on a message arrival, in a function to display TrayIcon, but happening only when compiled with geo-ip enabled. ??? ???
-
So my test is useless.. I'll try to disable geo-ip completely (as it is also useless for my usage) and I'll wait if the other kind of crash will appear or not.
-
By the way, just so you know, I did not abandon this topic. I stopped posting because I do not have much to say. I downloaded SVN the 5th of March, compiled with debugging (and geo-ip), and I'm running it inside gdb. No crash has occured since then, even though I reproduced the same situation a couple of times (popular files with many sources). I like the messaging theory though I do not have any facts to back it up.
-
I'm also using the same amule 20080309 version compiled without geoip inside gdb, but it didn't crashed.
Now I would like to know what does make amuled consuming all the cpu after some time of activity.. but this is another issue.....
-
By the way, just so you know, I did not abandon this topic. I stopped posting because I do not have much to say. I downloaded SVN the 5th of March, compiled with debugging (and geo-ip), and I'm running it inside gdb. No crash has occured since then, even though I reproduced the same situation a couple of times (popular files with many sources). I like the messaging theory though I do not have any facts to back it up.
Could you please try a little test, just to see if the messaging theory happens to anyone else.
Run two instances of amule with two different users (if not, it will complain saying that an instance is already running). Then, in the Messages tab, add to the second aMule the first one as a friend (right click-> add fiend). Use as IP 127.0.0.1 (ie. localhost) and as port, the TCP port of the first amule. Once added, right click on it, and Send Message, and send any message to the first one. If you have compiled with geo-ip and you're not filtering messages, the first aMule should crash in the SetTrayIcon function.
Although, I'm reading again this topic, and this particular crash (in SetTrayIcon) didn't happen to you, just to emisca...
-
I've disabled any message filter on the first amule, started another amule as another user, but I don't receive messages, either using 127.0.0.1 ip and the amule tcp port nor the public ip/port of the first amule....
-
I did exactly what kuanto said and I receive the messages and no crash.
-
Thanks for trying ;)
I've disabled any message filter on the first amule, started another amule as another user, but I don't receive messages, either using 127.0.0.1 ip and the amule tcp port nor the public ip/port of the first amule....
Any signal of the message? Neither a "message filtered" in the log, neither crash, neither the message itself...??
Maybe it wasn't succesfully sent from the second amule: Did the message you typed appear in the chat window of the sender?
I did exactly what kuanto said and I receive the messages and no crash.
Well. in my machine it crashes whenever I follow those steps. Maybe it happens only on some specific configuration: If it helps, my system is:
AMD64
Kubuntu 7.10
g++ (GCC) 4.1.3 20070929 (prerelease) (Ubuntu 4.1.2-16ubuntu2)
Linux kernel 2.6.22-14-generic
wxGTK 2.8.7 compiled from sources following wiki instructions
aMule CVS 23-03-2008 (just downloaded today version to try, and crashed too) compiled with
./configure --disable-optimize --enable-debug --enable-geoip --with-wx-config=/home/yo/usr/lib/wx/config/gtk2-unicode-release-2.8
Any idea? If you want me to run some test, just let me know ;)
-
My ipfilter list filtered it :-(
I've disabled it, and also "filter lan ip" but nothing to do...
-
My ipfilter list filtered it :-(
I've disabled it, and also "filter lan ip" but nothing to do...
Mmm, now that you mention it... 127.0.0.1 is in my ipfiler and I have "always filter LAN IPs" on. That message packet should be filtered from the beginning, shouldn't it?
-
Yes. Your amule doesn't filter it?
-
Yes. Your amule doesn't filter it?
Yes. The message is accepted, although it comes from an IP which is in ipfilter, and even that IP (127.0.0.1) should be filtered by the "Always filter LAN IPs" option.
Could this be another bug?
-
Some news.
I downloaded the Festor pre-compiled aMule package from here (to see if i was compiling it wrongly) : http://forum.amule.org/index.php?topic=13700.0 and I crashed it from another instance of aMule, sending a message. But this time, the backtrace says nothing about TrayIcon.
----------------------------=| BACKTRACE FOLLOWS: |=----------------------------
Current version is: aMule SVN using wxGTK2 v2.8.4 (Snapshot: Sat Mar 15 07:01:47 CET 2008)
Running on: Linux 2.6.22-14-generic x86_64
[2] wxThreadHelperThread::~wxThreadHelperThread() in amule [0x4486b5]
[3] wxFatalSignalHandler in /usr/lib/libwx_baseu-2.8.so.0[0x2ad75afd764c]
[4] ?? in /lib/libpthread.so.0 [0x2ad759a82100]
[5] wxMenuItemList::~wxMenuItemList() in amule [0x5768d8]
[6] wxDataObjectSimple::~wxDataObjectSimple() in amule [0x51ab27]
[7] wxDataObjectSimple::~wxDataObjectSimple() in amule [0x51b282]
[8] wxEvtHandler::ProcessEventIfMatches(wxEventTableEntryBase const&, wxEvtHandler*, wxEvent&) in /usr/lib/libwx_baseu-2.8.so.0[0x2ad75afd366f]
[9] wxEventHashTable::HandleEvent(wxEvent&, wxEvtHandler*) in /usr/lib/libwx_baseu-2.8.so.0[0x2ad75afd380f]
[10] wxEvtHandler::ProcessEvent(wxEvent&) in /usr/lib/libwx_baseu-2.8.so.0[0x2ad75afd3959]
[11] wxTimerBase::Notify() in /usr/lib/libwx_gtk2u_core-2.8.so.0[0x2ad75a959d76]
[12] ?? in /usr/lib/libwx_gtk2u_core-2.8.so.0 [0x2ad75a850a14]
[13] ?? in /usr/lib/libglib-2.0.so.0 [0x2ad75d69470b]
[14] g_main_context_dispatch in /usr/lib/libglib-2.0.so.0[0x2ad75d693fd3]
[15] ?? in /usr/lib/libglib-2.0.so.0 [0x2ad75d6972dd]
[16] g_main_loop_run in /usr/lib/libglib-2.0.so.0[0x2ad75d6975ea]
[17] gtk_main in /usr/lib/libgtk-x11-2.0.so.0[0x2ad75dc93883]
[18] wxEventLoop::Run() in /usr/lib/libwx_gtk2u_core-2.8.so.0[0x2ad75a84850d]
[19] wxAppBase::MainLoop() in /usr/lib/libwx_gtk2u_core-2.8.so.0[0x2ad75a8d9c8b]
[20] wxEntry(int&, wchar_t**) in /usr/lib/libwx_baseu-2.8.so.0[0x2ad75af7280c]
[21] std::basic_string<char, std::char_traits<char>, std::allocator<char> > std::operator+<char, std::char_traits<char>, std::allocator<char> >(std::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) in amule [0x516062]
[22] __libc_start_main in /lib/libc.so.6[0x2ad75be20b44]
[23] CryptoPP::BufferedTransformation::ChannelPut2(std::string const&, unsigned char const*, unsigned long, int, bool) in amule[0x4478f9]
--------------------------------------------------------------------------------
Could this be something Ubuntu specific?
Hope it helps ;)
-
Current version is: aMule SVN using wxGTK2 v2.8.7 (Snapshot: Mon Mar 31 07:01:45 CEST 2008)
Running on: Linux 2.6.18-6-686 i686
[2] ?? in amule [0x808b5b5]
[3] wxFatalSignalHandler in /usr/lib/libwx_baseu-2.8.so.0[0xb7469bd6]
[4] ?? in [0xb7f03420]
[5] ?? in [0x8d910b5]
[6] ?? in /usr/lib/libgdk-x11-2.0.so.0 [0xb6ce8b3f]
[7] ?? in /usr/lib/libglib-2.0.so.0 [0xb6a8a45d]
[8] g_main_context_dispatch in /usr/lib/libglib-2.0.so.0[0xb6a567e8]
[9] ?? in /usr/lib/libglib-2.0.so.0 [0xb6a59a3e]
[10] g_main_loop_run in /usr/lib/libglib-2.0.so.0[0xb6a59dc7]
[11] gtk_main in /usr/lib/libgtk-x11-2.0.so.0[0xb6e99cd4]
[12] wxEventLoop::Run() in /usr/lib/libwx_gtk2u_core-2.8.so.0[0xb762af0c]
[13] wxAppBase::MainLoop() in /usr/lib/libwx_gtk2u_core-2.8.so.0[0xb76cd23e]
[14] wxAppBase::OnRun() in /usr/lib/libwx_gtk2u_core-2.8.so.0[0xb76cc871]
[15] wxEntry(int&, wchar_t**) in /usr/lib/libwx_baseu-2.8.so.0[0xb73fa2ca]
[16] wxEntry(int&, char**) in /usr/lib/libwx_baseu-2.8.so.0[0xb73fa377]
[17] ?? in amule [0x824be9e]
[18] __libc_start_main in /lib/i686/cmov/libc.so.6[0xb70e4450]
[19] ?? in amule [0x8085991]
now i'm running with message filtering enabled. if it'll crush again, i'm going to make backtrace from wiki
Running Debian etch
-
thanks for the solution (the one 'bout turning message filtering on). seems like it helped - at least more than 24h without crashes ;)