[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Dchub-dev] [patch #2312] [Patch #2312] patch for bug 6475 - invalid $SR
From: |
noreply |
Subject: |
[Dchub-dev] [patch #2312] [Patch #2312] patch for bug 6475 - invalid $SR calls |
Date: |
Sun, 23 Nov 2003 07:38:57 -0500 |
User-agent: |
Mozilla/5.0 (compatible; Konqueror/3; Linux) |
Patch #2312 has been updated.
Project:
Category: None
Status: Open
Summary: patch for bug 6475 - invalid $SR calls
Follow-Ups:
Date: Wed 11/19/03 at 08:46
By: hhgoth
Comment:
The usage of the GString org after using the GLUS_SEND_TO_A_NICK doesn't seem
to be 100% reliable.
I encountered around 10 of 1000 invalid $SR-results sent to the event handler
inside an 100users hub.
...this way it seems to be stable
-------------------------------------------------------
Date: Wed 11/19/03 at 09:13
By: hhgoth
Comment:
maybe other events have to be checked if such a fix has to be implemented there
too.
-------------------------------------------------------
Date: Thu 11/20/03 at 18:32
By: ericprev
Comment:
you should have a memory leak because org_dup memory is not released no ?
-------------------------------------------------------
Date: Thu 11/20/03 at 18:47
By: hhgoth
Comment:
org_dup is a char no gstring - but put an free later on if you like ;)
-------------------------------------------------------
Date: Thu 11/20/03 at 18:57
By: hhgoth
Comment:
if freeing is impossible may it would be better to use an GString instead...
at least this is working stable in my hub with 140 users
I encountered not 1 error since I changed to this version.
before the hub crashed when activating the +slots command in the plugin
-------------------------------------------------------
Date: Sat 11/22/03 at 08:48
By: ericprev
Comment:
char * or gstring must be free. I suspect something inside GLUS_SEND_TO_A_NICK
to release the allocated memory too early. IMHO, the problem should come from
perl support. I think when perl functions return, they free the memory. If
there is no other plugins (perl/python or plugin), everything works fine but if
a plugin uses this memory later... boom.
-------------------------------------------------------
Date: Sat 11/22/03 at 11:02
By: hhgoth
Comment:
however - the problem has to be fixed for this and all other events this may
occur as well, or we'll run into errors we can't handle without massive problems
e.g. if a plugin fails - the hub will crash
...your fixes for the plugins are great eric :)
-------------------------------------------------------
Date: Sun 11/23/03 at 12:38
By: hhgoth
Comment:
added - the missing 'free(org_dup)' to patch (memory-leak)
...to implement until the 'real' bug is found
-------------------------------------------------------
-------------------------------------------------------
For more info, visit:
http://savannah.nongnu.org/patch/?func=detailpatch&patch_id=2312&group_id=3963
_______________________________________________
Message sent via/by Savannah
http://savannah.nongnu.org/
- [Dchub-dev] [patch #2312] [Patch #2312] patch for bug 6475 - invalid $SR calls, noreply, 2003/11/19
- [Dchub-dev] [patch #2312] [Patch #2312] patch for bug 6475 - invalid $SR calls, noreply, 2003/11/19
- [Dchub-dev] [patch #2312] [Patch #2312] patch for bug 6475 - invalid $SR calls, noreply, 2003/11/19
- [Dchub-dev] [patch #2312] [Patch #2312] patch for bug 6475 - invalid $SR calls, noreply, 2003/11/20
- [Dchub-dev] [patch #2312] [Patch #2312] patch for bug 6475 - invalid $SR calls, noreply, 2003/11/20
- [Dchub-dev] [patch #2312] [Patch #2312] patch for bug 6475 - invalid $SR calls, noreply, 2003/11/20
- [Dchub-dev] [patch #2312] [Patch #2312] patch for bug 6475 - invalid $SR calls, noreply, 2003/11/22
- [Dchub-dev] [patch #2312] [Patch #2312] patch for bug 6475 - invalid $SR calls, noreply, 2003/11/22
- [Dchub-dev] [patch #2312] [Patch #2312] patch for bug 6475 - invalid $SR calls,
noreply <=