mldonkey, a multi-networks file-sharing client - Bugs: bug #9138, /kernel: pid xxxx (mlnet-real),...
You are not allowed to post comments on this tracker with your current authentication level.
bug #9138: /kernel: pid xxxx (mlnet-real), uid 1002: exited on signal xx (core dumped)
Submitter: | None | ||
Submitted: | Fri 28 May 2004 08:58:35 AM UTC | ||
Category: | Core | Severity: | 3 - Normal |
Item Group: | Program malfunction | Status: | Works For Me |
Assigned to: | None | Open/Closed: | Closed |
Release: | 2.7.3 | Release: | |
Operating System: | FreeBSD | Binaries Origin: | CVS / Self compiled |
CPU type: | Intel x86 |
Jump to the original submission
Thu 07 Dec 2006 09:33:33 AM UTC, comment #52: |
raXion <raXion> |
Wed 06 Dec 2006 09:44:15 PM UTC, comment #51: Please check if this bug still is in current CVS version. |
spiralvoice <spiralvoice> |
Tue 14 Nov 2006 08:29:43 PM UTC, comment #50: What happens when you compile MLDonkey without
|
spiralvoice <spiralvoice> |
Tue 07 Nov 2006 09:18:33 PM UTC, comment #49: Please check if this bug still is in current CVS version.
|
spiralvoice <spiralvoice> |
Sun 16 Jul 2006 12:56:15 AM UTC, comment #48: Is this bug still valid?
|
spiralvoice <spiralvoice> |
Mon 13 Feb 2006 02:02:00 PM UTC, comment #47: The gdb-backtrace seems to be the same:
|
raXion <raXion> |
Mon 13 Feb 2006 09:44:15 AM UTC, comment #46: I recompiled MLDonkey with the following options (hoping this will produce better backtraces):
|
raXion <raXion> |
Mon 13 Feb 2006 09:35:32 AM UTC, comment #45: I did, and I also get core dumps ;-(.
|
raXion <raXion> |
Sun 12 Feb 2006 03:59:55 PM UTC, comment #44: Did you ever try if libpthread produces the same core dump? |
spiralvoice <spiralvoice> |
Sun 12 Feb 2006 03:04:17 PM UTC, comment #43: MLDonkey was made using the following command-line:
|
raXion <raXion> |
Sat 11 Feb 2006 12:27:09 AM UTC, comment #42: um i can't read anything usefull in this backtrace. was mldonkey compiled with debugsymbols ( --enable-debug )? |
Amorphous <amorphous> |
Wed 08 Feb 2006 08:22:12 AM UTC, comment #41: I used gdb with the generated core-file 'gdb -c mlnet-real.core /usr/local/bin/mlnet-real':
|
raXion <raXion> |
Sun 22 Jan 2006 12:23:21 AM UTC, comment #40: please get a backtrace. after you get "Program received signal SIGSEGV..." type in the command bt. |
Amorphous <amorphous> |
Fri 20 Jan 2006 03:54:48 PM UTC, comment #39: Hi, here is what I've done and what output I got. It doesn't seem to say very much to me. Did I used gdb right at all?
|
raXion <raXion> |
Fri 20 Jan 2006 02:07:49 PM UTC, comment #38: I still have random core dumps of mldonkey.
|
raXion <raXion> |
Mon 21 Nov 2005 09:15:27 PM UTC, comment #37: Please check if this bug still is in current CVS version.
|
spiralvoice <spiralvoice> |
Sat 23 Jul 2005 12:58:32 PM UTC, comment #36: Does that patch print out the function backtrace when the application does an segv? The only thing you need to do is add b to that env. variabl?
|
Amorphous <amorphous> |
Fri 22 Jul 2005 10:39:49 AM UTC, comment #35: Thanks.
|
raXion <raXion> |
Thu 21 Jul 2005 03:54:13 PM UTC, comment #34: Maybe this can help: "Patch for native function backtraces"
|
spiralvoice <spiralvoice> |
Tue 19 Jul 2005 06:13:03 PM UTC, comment #33: You could try to use gdb on mldonkey (and try to get an backtace), but i can't tell you from my head how to do that. Perhaps someone else can say how to exactly do that. Otherwise i will try, when i have time for mldonkey again. |
Amorphous <amorphous> |
Tue 19 Jul 2005 05:28:07 PM UTC, comment #32: I guess you did it the right way. It seems it is not possible to get an backtrace from an segv-ing application in that way. Perhaps the segv occurs in some c-code and not in the ocaml part? I hope someone else has some more insight how to procede here. |
Amorphous <amorphous> |
Tue 19 Jul 2005 08:56:59 AM UTC, comment #31: After crashing I can only kill the debugger to get out. It makes no difference wheter I enter 'y' or 'n' in the debugger after crash: I only get the same question every time. (See the end of my last post.)
|
Anonymous |
Fri 10 Jun 2005 11:16:14 AM UTC, comment #30: I couldn't get the backtrace!
|
raXion <raXion> |
Tue 07 Jun 2005 02:57:08 PM UTC, comment #29: I started wit ocamldebug and it seems to work. Only your option "printf '-l %p'" doesn't work. Is this critical for debugging? Here's the output I got:
|
raXion <raXion> |
Sun 05 Jun 2005 03:57:04 PM UTC, comment #28: You get mlnet.byte by the following command after you did run configure:
|
Amorphous <amorphous> |
Sun 05 Jun 2005 03:27:04 PM UTC, comment #27: Where did I find mlnet.byte?
|
raXion <raXion> |
Fri 03 Jun 2005 02:49:42 PM UTC, comment #26: copy mlnet.byte to your mldonkey directory and cd into it
|
Amorphous <amorphous> |
Thu 02 Jun 2005 03:23:47 PM UTC, comment #25: Here's my output of make.
|
Anonymous |
Wed 01 Jun 2005 08:01:30 AM UTC, comment #24: Could you tell us the output of buildinfo ?
|
Amorphous <amorphous> |
Tue 31 May 2005 10:32:03 AM UTC, comment #23: I had six core-dumps since my last post, 5 days ago. |
raXion <raXion> |
Thu 26 May 2005 10:19:01 AM UTC, comment #22: The recompiled binary is running now ...
|
raXion <raXion> |
Tue 24 May 2005 03:16:42 PM UTC, comment #21: Please try compiling with --enable-checks
|
spiralvoice <spiralvoice> |
Tue 24 May 2005 10:54:51 AM UTC, comment #20: I still get those core dumps with the following environment:
|
raXion <raXion> |
Wed 16 Mar 2005 02:22:19 AM UTC, comment #19: does this still happen? |
Amorphous <amorphous> |
Fri 01 Oct 2004 02:03:10 PM UTC, comment #18: 'ulimit' says 'umlimited', if started as my MLD-user.
|
raXion <raXion> |
Fri 01 Oct 2004 01:09:14 PM UTC, comment #17: Use ulimit |
spiralvoice <spiralvoice> |
Wed 29 Sep 2004 11:42:39 AM UTC, comment #16: I have enabled this networks: Overnet, Donkey, Bittorrent, Fastrack and FileTP. But I got 5 core-dumps during the last 2 days, all after 3 to 4 hours after starting the mlnet-core.
|
raXion <raXion> |
Thu 19 Aug 2004 12:56:22 PM UTC, comment #15: Please pay also attention to the RAM limits for the MLDonkey process imposed by ulimit. |
spiralvoice <spiralvoice> |
Thu 19 Aug 2004 12:03:13 PM UTC, comment #14: Today I updated to 2.5.28. |
raXion <raXion> |
Tue 17 Aug 2004 11:15:16 AM UTC, comment #13: I changed ulimit settings according to http://support.zeus.com/faq/zlb/v1/entries/os/freebsd.html . I have autoconf-2.59 installed ... I also disabled both Gnutella-Networks. |
raXion <raXion> |
Sun 15 Aug 2004 03:51:49 PM UTC, comment #12: Please check ulimit settings for mlnet, they tend to be pretty tight on FreeBSD, also install autoconf-2.59 to compile the core. |
spiralvoice <spiralvoice> |
Wed 11 Aug 2004 04:43:42 PM UTC, comment #11: I would like to, but this depends on bugs #9795 for me. |
raXion <raXion> |
Tue 10 Aug 2004 11:04:12 AM UTC, comment #10: Please check with current CVS versions, also please check whether there is a small ulimit for MLDonkey regarding memory. |
spiralvoice <spiralvoice> |
Tue 27 Jul 2004 01:00:56 PM UTC, comment #9: I have 1536 MB of RAM installed and there is always free memory available.
|
Anonymous |
Thu 22 Jul 2004 11:25:45 PM UTC, comment #8: Please check whether MLDonkey uses too much RAM and dies when no free RAM is available anymore. What networks are enabled and used? Some of them have known memory leaks (Gnutella for example), take a look at the other bug reports. |
spiralvoice <spiralvoice> |
Thu 01 Jul 2004 12:16:29 PM UTC, comment #7: Today I ran Memtest v3.0 from Boot-CD and the standard test run (took about 90 minutes) didn't report any errors.
|
Anonymous |
Thu 17 Jun 2004 12:27:07 PM UTC, comment #6: CPU fan is working.
|
Anonymous |
Sat 05 Jun 2004 07:19:58 AM UTC, comment #5: Try this for memory test:
|
Anonymous |
Thu 03 Jun 2004 10:09:52 PM UTC, comment #4: Good to hear that. My CPU is not overclocked, but maybe I should try some more conservative memory speed settings. It's set to the default "normal", I think, but I will check this ... |
Anonymous |
Wed 02 Jun 2004 09:05:54 PM UTC, comment #3: This signals mostly related to bad memory or overclocking. I use the same OS but on a good old Celeron 300A without any probs.
|
Artifex Maximus <artifex> |
Mon 31 May 2004 10:08:47 PM UTC, comment #2: I've attached the output of `grep mlnet-real` /var/log/messages` as mlnet-dies.txt. It seems that mlnet-real dies at least once a day. Currently we have a total of 35 files in download, in sum 20 GB.
|
Anonymous |
Sat 29 May 2004 06:07:40 PM UTC, comment #1: Do you know under which circumstances the core dies? How long did the core run? |
spiralvoice <spiralvoice> |
Fri 28 May 2004 08:58:35 AM UTC, original submission:
From time to time I get one of the two messages after mlnet-real died:
|
Anonymous |
Depends on the following items: None found
Items that depend on this one: None found
There are 0 votes so far. Votes easily highlight which items people would like to see resolved in priority, independently of the priority of the item set by tracker managers.
Follow 14 latest changes.
Date | Changed by | Updated Field | Previous Value | => | Replaced by |
---|---|---|---|---|---|
2006-12-07 | spiralvoice | Status | Need Info | Works For Me | |
Open/Closed | Open | Closed | |||
2006-06-23 | spiralvoice | Release | 2.5.30.15 | 2.7.3 | |
2005-07-19 | beedauchon | Status | None | Need Info | |
2005-06-02 | None | Attached File | - | Added make.txt, #2560 | |
2005-05-24 | spiralvoice | Status | Fixed | None | |
Open/Closed | Closed | Open | |||
Release | 2-5-21 | 2.5.30.15 | |||
Release | 2.5.-21 | ||||
2005-04-12 | spiralvoice | Open/Closed | Open | Closed | |
2005-03-16 | amorphous | Status | None | Fixed | |
2004-07-29 | raXion | Carbon-Copy | - | Added raxion | |
2004-05-31 | None | Attached File | - | Added mlnet-dies[1].txt, #1360 | |
2004-05-28 | None | Attached File | - | Added part_of_my_messages.txt, #1348 |
Powered by Savane 3.13-758e.
Corresponding source code
Since several weeks I only use BT-module and had no crashes. The version I've installed is from ports, which is synced with the CVS version from 2006/11/21.
I compiled my last upgrades WITHOUT_THR_THREAD_MODEL.
As I haven't used the edonkey-module in the last time, I can't say if the crashes are gone forever now. I would propose to close this bug-report for now. I will try to reopen it, if this would be neccessary.
Greetings,
raXion