[
https://issues.apache.org/jira/browse/LOGCXX-458?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14993282#comment-14993282
]
Thorsten Schöning commented on LOGCXX-458:
------------------------------------------
The changes in LOGCXX-394 have nothing to do with your leak, I just thought of
that bug after reading your comment on SO mentioning exactly the changes I
committed for LOGCXX-394. But your 0.10.0 shouldn't contain those changes
unlees some maintainer backported them what I really doubt. So yes, it looks
like the leak is related to "static LevelPtr level(...)"... I have no
experience with Valgrind, but may it be that Valgrind is recognizing those
memory leaks just too early and that because of static the LevelPtr is
destructed after Valgrind can recognize that?
In any case, you can try with a current trunk HEAD, which contains LOGCXX-394,
to see if that changes things. The function statics made problems in multi
threading apps anyways and where therefore changed.
> Does log4cxx::Level::getError() leak memory?
> --------------------------------------------
>
> Key: LOGCXX-458
> URL: https://issues.apache.org/jira/browse/LOGCXX-458
> Project: Log4cxx
> Issue Type: Bug
> Affects Versions: 0.10.0
> Reporter: Michael Herzlich
>
> Hi there,
> I have posted a question about the possible memory leak at stackoverflow:
> http://stackoverflow.com/questions/33546286/does-log4cxxlevelgeterror-leak-memory?noredirect=1#comment54872303_33546286
> Could anybody verify this? Is this behavior by intention?
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
