Jakub Jankowski <shasta@toxcorp.com> writes:
On Thursday 22 of March 2012, Peter Czanik wrote:
On 03/22/2012 02:46 PM, Jakub Jankowski wrote:
I know at least about one memory leak that has been once fixed, but then reintroduced between 3.3.2 and 3.3.3, and I think is present since then. I've mentioned it on this list in February: https://lists.balabit.hu/pipermail/syslog-ng/2012-February/018334.html
It would be really nice if developers could take a look at the code again. :)
Check this: https://bugzilla.balabit.com/show_bug.cgi?id=160 The mentioned patch is available at http://git.balabit.hu/?p=bazsi/syslog-ng-3.3.git;a=commit;h=3c14a264ab4c76b9...
Unfortunately, this doesn't fix the memleak I'm seeing. I've just recreated my testbed and tried current 3.3 HEAD (using Gergely's tarball as source: http://packages.madhouse-project.org/syslog-ng/3.3/3.3.4/syslog-ng-3.3.4-201...) and can confirm "my" leak is still happening.
Thanks for the detailed report, I was able to reproduce the problem (at least the leak - I didn't get any empty files), and am now looking into it. -- |8]