I've been looking through the archives for information, but I haven't stumbled on a solution to my problem.
I'm running ds-389 (389-ds-base-126.96.36.199) on a centos 7 box (CentOS Linux release 7.2.1511). I have a centos OS client configured using SSL/TLS
which queries the LDAP server. Per a previous thread, I configured the memeberOf plugin and all seems to be working properly.
I have a php script that will run on the client and change the LDAP password for the user. The problem is, the script looks for the SSHA has
of the password when an ldapsearch is issued.
However, when I issue a general ldapsearch (anonymously) I don't get the userpassword field. I read in your archives that I might have
to be the "directory manager" user in order to see the hashed password. I've been playing around with the ldapsearch syntax, but I can't
quite get it right.
Anyway, my question is, can I set a flag in 389-ds that will display the hashed userpassword? I think that will solve my problem with the php script returning an error that it can't retrieve the old password.
We are implementing PassSync 1.1.7 in a Windows Server 2012R2 domain and I am not seeing any indication the passsync.log file that passwords are being sent to the 389 DS. I have confirmed that the PassSync service starts without an error on the two Active Directory domain controllers, and the OU structure and user accounts have synced between Windows and 389. At this point, it just not sending password changes to the 389 DS. I was wondering if there was documentation that described any required security settings within the Windows domain? I am afraid that we have enabled some security setting in a GPO.
I just discovered that I have a mismatch of BER sizes (nsslapd-maxbersize) between some of my masters. On the masters that we consider the authorative ones, the BER size is (cough, cough) 209M while in the rest of the environment is the default 2M. I do see occasional errors on the 2M masters complaining and suggesting to increase the BER size. A couple of questions:
1. If the BER size of the supplier exceeds that of the consumer, will it be automatically split into smaller chunks or will the update just fail without any retries?
2. Should I match the BER size accross the environment? I think it’s obvious, but still asking just in case.
Since today is the day for announcements from the project ... we have
new command line tools! These will be shipped in F28, and our perl
tools will be removed in F29.
We have been hard at work on new cli tools for the project to replace
our perl scripts. Our tools are all based on the lib389 python
framework. We have three tools that we will ship:
Each tool has a different responsibility.
This tool will administer all local aspects of a system. It will
generally require root/dirsrv permissions. It's used for
starting/stopping the server, running local tasks like db2index,
db2ldif, ldif2db and others.
This tool administers an "online" instance. It's the tool that needs
cn=Directory Manager permissions, so it's focused on cn=config and
changes. Examples are configuring plugins, creating backends, adding
indexes, and checks for server configuration sanity.
This tool manages the content of a backend - and can even be used for
self service. This will populate sample data into a backend, it can
manage users, groups, ous, and more. Power users could even use it to
self enroll ssh keys or change their details in some organisations.
Each tool will come with extensive help options, and as we get closer,
we'll write more to help explain the new tools.
If you are interested in using these tools or contributing, please
contact us for more information,
Red Hat, Australia/Brisbane
In Fedora 28 (389-ds-base-1.4.0) we are deprecating the
389-console/Admin Server. Instead we will be offering a new web UI via
a Cockpit plugin to handle the Directory Server Administration. See
Why Cockpit? Well Cockpit has its pros & cons, but since it has
built-in functionality that solves many problems for us we decided to
use it instead of developing a full-blown standalone web application
that needs run on Apache.
- Authentication is built into Cockpit. However, you must login as a
system user, and one with superuser rights, in order to administer the
server. You will not log in with an LDAP entry. Authentication to the
Directory Server is then done via LDAPI/autobind. See
- Remote server administration. You can register other systems into
Cockpit. Cockpit then establishes ssh sessions with those remote
systems. This allows us to securely manage all your Directory Servers
from a single location.
- Lightweight. The new UI will no longer need things like
"o=netscaperoot", or a full blown http server. It will be very
lightweight and it will run inside of Cockpit (which is also VERY
lightweight). However, this also means if you are on a platform that
does not offer Cockpit (like Solaris/HPUX) then unfortunately you will
not have a console/UI available to you.
- We can not use fancy UI frameworks like django/jinja in cockpit. So
the UI will be simple - that's not really a big con, but it does limit
what functionality we can easily add.
- Authentication. You must use a system user (root or a user with
superuser privileges), not an LDAP account/DN, but you can use
autobind/LDAPI to map a system user to an ldap account.
Note - there will NOT be an LDAP browser in the new web UI. However, we
will be accepting upstream source contributions towards an LDAP
browser, but as of right right now we just do not have the resources to
add a nicely polished LDAP tree browser. Perhaps in a future release
we can do it, but it is considered a low priority. In the meantime for
F28 (389-ds-base-1.4.0), there are other free third-party LDAP browsers
out there that you can use, we just won't have one built into the UI at
As development continues on the new "web console" we will be providing
alpha/beta releases for review/feedback. And your feedback is
important! Expect to start seeing these "test releases" in a few
months, and installing Cockpit and adding the DS UI plugin is very
simple and only takes a few seconds.
If you have any questions/requests please let us know.
We have recently notice that there is a significant fault in thread
safety of the i686 platform of 389 Directory Server. ONLY the i686
version is affected, all other versions are safe.
Large parts of the servers thread safety are affected including
reference counting, monitor counters, and io-event notification and
safety. The fault may cause unbounded memory leaks, crashes,
inconsistent data reporting and more. This fault has been present for
at least 6 months or more.
However, we have received no reports from users related to this error
As a result, we suspect no one is using i686 anymore.
Given the extensive effort to fix this, combined with the short future
life of i686 as a whole (given topics like 2038 problem), and the fact
fedora stopped supporting server i686 last year, we would like to
remove our i686 support from the project.
It is likely we will target this for 1.4.x series of Directory Server,
and retroactively for 1.3.x.
Red Hat, Australia/Brisbane