Summary | Inconsistencies between datatree and LDAP when working on a Kolab backend |
Queue | Kolab |
Type | Bug |
State | Resolved |
Priority | 2. Medium |
Owners | stuart (at) |
Requester | a.gungl (at) gmx (dot) de |
Created | 02/07/2005 (7442 days ago) |
Due | |
Updated | 11/20/2006 (6791 days ago) |
Assigned | 11/22/2005 (7154 days ago) |
Resolved | 11/20/2006 (6791 days ago) |
Github Issue Link | |
Github Pull Request | |
Milestone | |
Patch | No |
State ⇒ Resolved
#4662.default with Kolab, that makes sure that we always use the user's mail
attribute internally.
I added an example hook to hooks.php.dist, but it is untested, and I
have no idea if it does what it is supposed to do. Please test.
with regard to the upcoming kolab 2.1 and multidomain support it would
be better to use the email address for internal queries.
Ciao,
Tobias
Queue ⇒ Kolab
matter which one the user uses for login. Which one makes more sense,
especially looking ahead to multidomain support in Kolab 2.1? How
will user names look like then?
can't say anything about that at the moment. Maybe later I'll have
some time to install it.
Hopefully somebody can post come comments....
matter which one the user uses for login. Which one makes more sense,
especially looking ahead to multidomain support in Kolab 2.1? How will
user names look like then?
"mail" as the internal user name instead of "uid"? Does this have any
advantages regarding to Kolab?
same issue as described in the ticket (tested with both the latest
Horde-stable and Horde-cvs). The problem appears only when "uid" isn't
the same as "mail". So in this case Horde with Kolab backend works
correctly _only_ if you login with your mail address, but not with
uid! Of course, I don't think we should use mail instead of uid as
internal user name. Would be the best if we can use both of them
without problems.
State ⇒ Feedback
Priority ⇒ 2. Medium
"mail" as the internal user name instead of "uid"? Does this have any
advantages regarding to Kolab?
State ⇒ Assigned
Priority ⇒ 3. High
Type ⇒ Bug
Summary ⇒ Inconsitences between datatree and LDAP when working on a Kolab backend
Queue ⇒ Horde Framework Packages
State ⇒ Unconfirmed
This works fine.However, the queries in the datatree are done based on
the login value of the user. If uid=mail, that doesn't matter, but
having uid=name and mail=name@domain might let the datatree double the
data for the same user. In my case there is the problemthat data for
the user can't be retrieved when he logs in via uid, while everything
works fine when using the mail value.
There has been a related ticket some days ago, which was about reading
and storing in LDAP. This _is_ solved. This ticket is about using the
same approach in the datatree, i.e. use the value for the primary LDAP
search attribute (Prefs/Prefs/kolab.php sets $params['uid'] =
array('mail', 'uid'); to access the Kolab LDAP server) for the
datatree access (in this example the value of "mail" and not of "uid").