After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 116162 - Memory leak somewhere in gnet_inetaddr_new
Memory leak somewhere in gnet_inetaddr_new
Status: VERIFIED INCOMPLETE
Product: gnet
Classification: Other
Component: general
unspecified
Other FreeBSD
: Normal critical
: ---
Assigned To: dhelder
dhelder
: 118095 (view as bug list)
Depends on:
Blocks:
 
 
Reported: 2003-06-28 01:30 UTC by chris
Modified: 2009-08-15 18:40 UTC
See Also:
GNOME target: ---
GNOME version: ---


Description chris 2003-06-28 01:30:03 UTC 
For information on how to reproduce this bug:

http://www.gnetlibrary.org/pipermail/gnet-dev/2003-May/000244.html

Also, if using the above example you might want to change getc() to sleep
(1) and let it run for a few minutes to see the impact this can have on a 
program with hundreds/thousands of short-term connections, I had to paste 
in about 1000-2000 characters to really be sure this was it.

Personally, I don't use the same function as in the above example, but by 
using gnet_tcp_socket_connect_async() the same problem arises (because 
somewhere along the lines they both run the same peice of code.)

I'll be tracking this down over the weekend myself, but I think somebody 
thorougly familiar with the gnet code should take a long hard look at it.
Comment 1 dhelder 2003-06-30 23:11:23 UTC 
There is a bug in gnet_inetaddr_unref, where it does not get deleted
properly.  This has been fixed in HEAD.

What version of GNet are you using?
Comment 2 chris 2003-06-30 23:16:07 UTC 
FreeBSD: gnet-2.0.3, from source
Linux: 2.0.3 (debian package (rev 1))
Comment 3 chris 2003-06-30 23:23:25 UTC 
What is HEAD, and where do I get it.  I need this fixed asap.
Comment 4 dhelder 2003-06-30 23:27:33 UTC 
HEAD = cvs

I looked at this last month when it was first reported (see response
on email list).

I would guess that either:
1. this is a problem in getaddrinfo()
2. this has been fixed by fixing gnet_inetaddr_unref.

Please try the CVS version.  I will make try to make a new release
this week too.

David
Comment 5 Charles Kerr 2003-07-23 05:39:29 UTC 
*** Bug 118095 has been marked as a duplicate of this bug. ***