Alan,
the code I transcripted is from the radius_ldap.c routine which is part of
the Cistron LDAP Patch downloaded from
http://works.agni.com/cistron-ldap.html .
I was not involved with the download of the our code, so maybe we are not
using the 'official' one?
Where is the rlm_ldap code you mentioned?
Thanks,
---
Juan
----- Original Message -----
From: <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Wednesday, August 29, 2001 1:58 PM
Subject: Re: (Freeradius 0.1 vs. Freeradius 0.2) + LDAP
> "Juan Marchionatto" <[EMAIL PROTECTED]> wrote:
> > Hi. I think I may have found a memory leak's cause in the LDAP routines.
>
> That sounds great!
>
> > The ldap_first_entry routine is reusing the variable 'result', which was
> > previously assigned on ldap_search_st, so any of the ldap_msgfree later
> > calls will not pass the original value, as it should.
>
> Hmm... from my reading of the ldap docs, that's what it's *supposed*
> to do. You first search for the results, and then find the first
> entry within that results.
>
> > If this is the problem, using a second (LDAPMessage *) variable for the
> > entry-retrieving calls, should do it.
> >
> > Just as an example, check this code fragment:
> >
> > if ((result = ldap_first_entry(ld_inst, result)) == NULL) {
> > ldap_msgfree(result);
> > return 0;
> > }
> >
> > It is wrong, because you are always passing NULL to ldap_msgfree instead
> > of the original value returned from ldap_first_entry
>
> Yes, that code is wrong. But I don't see any code like that in the
> current LDAP module.
>
> > It should be:
> >
> > if ((another_ptLDAPMessage_variable = ldap_first_entry(ld_inst,
> > result)) == NULL) {
> > ldap_msgfree(result);
> > return 0;
> > }
>
> Yes, that code is correct. And that's what the current rlm_ldap
> code does.
>
> Alan DeKok.
>
> -
> List info/subscribe/unsubscribe? See
http://www.freeradius.org/list/users.html
>
-
List info/subscribe/unsubscribe? See http://www.freeradius.org/list/users.html
