This section focuses on how to use LDAP as a NIS substitute
for user accounts management. Having a lot of user accounts on several hosts
often causes misalignments in the accounts configuration. LDAP can be used to
build a centralized authentication system thus avoiding data replication and
increasing data consistency.
At the moment the most used method to distribute users account data and
other information through a network is the Network Information Service (NIS).
Like LDAP, NIS is a distributed service that allows to have a central server
where configuration files such as passwd, shadow, groups, services, hosts etc.
are kept. The NIS server is queried by NIS clients to retrieve this information.
LDAP can offer the same functionality of NIS, moreover there are several
advantages on using LDAP:
Information on the LDAP server can be easily used for several
purposes. As outlined in this HOWTO, the same users entries on the LDAP database
can be used for other applications like phone directories, mail routing, staff
databases etc., thus avoiding data replication and inconsistency.
LDAP allows complex access control lists to be applied on the
database. This allows for a fine grain tuning of permissions on the database
A secure transmission channel between the LDAP server and the clients
can be implemented through the Secure Socket Layer (SSL).
A fault tolerant service can be implemented using slapd
replication  and DNS round robin queries
(this is not covered in this document).
Having a single instance of users on the network helps to
maintain users on many hosts from a single management point (i.e. you can
create and delete accounts in the LDAP server and this changes are available
immediately to LDAP clients).
Herein I'll focus on how an LDAP server can be used for authentication
and authorization on systems providing the Pluggable Authentication Module
(PAM) and the Name Service Switch (NSS) technologies, in particular I'll refer
to the Linux operating system even if this instructions can be applied to other
The environment proposed consists of an LDAP server where users account
data is stored in a convenient format and a set of Un*x clients using this
information to authenticate and authorize users on resources in a standard Un*x
A secure channel is also required in client/server communications since
critical information such as user account data, should not be sent in clear
over the network, this channel will be provided by the Secure Socket Layer.
On the client side a caching mechanism, needed for performance issues,
can be provided by the Name Service Caching Daemon.
All (almost) the software used to build this system is Open Source.
The Pluggable Authentication Module allows integration of various
authentication technologies such as standard UNIX, RSA, DCE, LDAP etc. into
system services such as login, passwd, rlogin, su, ftp, ssh etc. without
changing any of these services.
First implemented by Sun Solaris, PAM is now the standard authentication
framework of many Linux distributions, including RedHat and Debian. It provides
an API through which authentication requests are mapped into technology
specific actions (implemented in the so called pam modules). This mapping is
done by PAM configuration files, in which, for each service are basically given
the authentication mechanisms to use.
In our case, the pam_ldap module, implemented in the shared library
pam_ldap.so, allows user and group authentication using an LDAP service.
Each service that needs an authentication facility, can be configured
through the PAM configuration files to use different authentication methods.
This means that it is possible, using the PAM configuration files, to write a
custom list of requirements that an user must satisfy to obtain access to a
Once an user is authenticated, many applications still need access to
user information. This information is traditionally contained in text files
(/etc/passwd, /etc/shadow, and
/etc/group) but can also be provided by other name
As a new name service (such as LDAP) is introduced it can be implemented
either in the C library (as it was for NIS and DNS) or in the application that
wants to use the new nameservice.
Anyway, this can be avoided using a common, general purpose, name
service API and by demanding to a set of libraries the task of retrieving this
information performing technology based operations.
This solution was adopted in the GNU C Library that implements the
Name Service Switch, a method originated from the Sun C
library that permits to obtain information from various name services through a
NSS uses a common API and a configuration file
(/etc/nsswitch.conf) in which the name
service providers for every supported database are specified.
Using the nss_ldap shared library it is possible to implement the maps
above using LDAP, anyway here I'll focus only on the LDAP implementation of
shadow, passwd and group database tough all the maps above can be implemented.
For most of the other maps it is even unadvisable to store them in ldap, as they
tend not to change too often, so it is not a problem to have them locally as files,
and storing them in ldap would cause some minor performance loss.
For our application LDAP is used to provide clients with information
about user accounts and user groups. The standard objectclasses that are used
to represent users and groups are: top, posixAccount, shadowAccount and
Users entries on the database must belong at least to the top,
posixAccount and shadowAccount objectclasses. Group entries must belong to the
top and posixGroup objectclasses.
The implementation of pam_ldap and nss_ldap that we use refers to this objectclasses, that
are described in RFC 2307.
Note: Actually LDAP NSS recognize other objectclasses
The Name Service Caching Daemon (NSCD) is used to cache name service
lookups and can improve performance with the services provided by the NSS.
It must be tuned with a large cache for passwd entries in order to have
acceptable performance on the client side.
It has some disadvantages however, like the introduction of cache inconsistencies,
so you would want to be sure you need this before you use it. We have succesfully running
some systems without it, and personally i think that it isn't really neccesary on
relatively small systems.
SSL is needed in the communication between the LDAP server and the
clients libraries (pam_ldap.so and nss_ldap.so), since sensible data, such as
password entries, needs to be encrypted between the client and the server. SLL
also permits the client to uniquely identify the server, thus avoiding to
obtain authentication informations from an untrusted source.
Client authentication (the server identifies the client) is not supported
in the current implementation of pam_ldap and nss_ldap modules tough it may be
You can refer to the LDAP-HOWTO for instruction on installation and
configuration of LDAP
Once slapd is properly configured we need to insert some data for the
initial creation of the database. Therefore an LDIF (LDAP Data interchange
format) file must be created. This is a text file that can be imported in the
LDAP database with the command:
#ldif2ldbm -i your_file.ldif
Note: ldif2ldbm is provided with the OpenLDAP 1.2.x package, if you use
OpenLDAP 2.0.x, you should use the ldapadd command
(after the server is started).
If you use OpenLDAP 2.0.x (LDAPv3) you can find the standard nis schema
in the file etc/openldap/schema/nis.schema, include
it in your
slapd.conf with the include directive, to have schema
Here is an example of a minimal LDIF file. Each entry is
separated by a blank line.
dn:ou=groups, dc=yourorg, dc=com
dn:ou=people, dc=yourorg, dc=com
dn: cn=Giuseppe LoBiondo, ou=people, dc=yourorg, dc=com
cn: Giuseppe Lo Biondo
sn: Lo Biondo
gecos:Giuseppe Lo Biondo
dn: cn=mygroup, ou=groups, dc=yourorg, dc=com
Note: Note that lines that are too long are continued on the following
line started by a tab or a space, this in true too for LDIF format files
Here we defined the base DN for the orgazation dc=yourorg,
dc=com under which are contained two sub organizational units: people
and groups. Then is described a user that belongs to the people organizational
unit and a group (which the users belongs to) under the groups organizational
The LDIF file must be imported in the server while it is not running
since the ldif2ldbm command builds the database directly, bypassing the LDAP
server. Once the LDIF file is imported into the database, the server can be
On the client side pam_ldap.so and nss_ldap.so are required and they
must be compiled using the Netscape LDAP Library (Mozilla) since it provides
the required LDAPS (LDAP over SSL) API. The library is distributed in a binary
package under Netscape One license and is not open source (it is public domain
The package can be extracted, for example, in the directory
Client libraries must also have access to a certificate database
containing the LDAP (stunnel) server certificate and the CA certificate of the
CA that signed the server certificate (marked as trusted).
The certificate database must be in Netscape format since the Mozilla
LDAP API used to compile pam_ldap and nss_ldap uses certificate databases in
To deal with such certificate databases it is convenient to use the
certutil utility found in the PKCS#11 package provided by Netscape
The main configuration file for LDAP clients is
Note that if you use nss_ldap, you don't strictly need to use pam_ldap.
You can use the pam_unix_auth module instead, since nss_ldap maps all
getpw* and getsh* calls into LDAP lookups and pam_unix_auth uses this calls to
To compile and install pam_ldap, do the following:
$ ./configure --with-ldap-lib=netscape4 \
# make install
The configure switch --with-ldap-lib tells which LDAP library you are
going to use.
The switch --with-ldap-dir tells where you have installed your Netscape
This will install /lib/security/pam_ldap.so.1 and the
PAM has to be properly configured in order to access the new authentication
system. PAM configuration files are located in the directory
/etc/pam.d and are named after the service for which
authentication is provided.
For example this is the PAM configuration file for the login service (in a
file named login).
Standard PAM configuration files for use with PAM can be found in the
pam_ldap source distribution, in the directory
This files can be copied in the /etc/pam.d directory.
Caution must be given when performing this operation, since if something goes
wrong you probably will not be able to login again. It is suggested to make a
backup copy of /etc/pam.d before installing new files there
and to leave an open privileged shell.
Note: In the example pam.d directory,
a sshd file is not present, so unless you
create one, you will be unable to login via ssh, if it uses pam (OpenSSH does
After you've unpacked the sources, check the makefile. For most
configurations, it doesn't need to be edited. Anyway, if you want to use SSL
you must link against an SSL aware LDAP library, such as the Netscape one.
Assuming that the ldap sdk is in /usr/local/ldapsdk
you have to modify the Makefile to enable SSL. Look for NSFLAGS in
Makefile.linux.mozilla and uncomment -DSSL.
Also check the LIBS definition to see if the ldapssl library specified
in the file is the same that you have installed (ldap_nss.so compiles with both
libldapssl40 and libldapssl30).
Then you can install the library:
$ make -f Makefile.linux.mozilla
# make -f Makefile.linux.mozilla install
this installs /lib/libnss_ldap.so,
which is the nss_ldap library, and a set of example configuration files,
/etc/nsswitch.ldap and /etc/ldap.conf, in case they do not exist
Once you have installed it you must edit the NSS configuration file
/etc/nsswitch.conf. Tough LDAP can be used for all the
services we use it only for passwd, group and shadow therefore we should have
NSCD is already available in many Linux distributions, anyway it can be
found within the GNU C library package.
The NSCD configuration file is /etc/nscd.conf. Each
line specifies either an attribute and a value, or an attribute, cachename, and a
value. Fields are separated either by SPACE or TAB characters. cachename can be
hosts, passwd, or groups (in our case we won't cache hosts).
enable-cache passwd yes
positive-time-to-live passwd 600
negative-time-to-live passwd 20
suggested-size passwd 211
keep-hot-count passwd 20
check-files passwd yes
enable-cache group yes
positive-time-to-live group 3600
negative-time-to-live group 60
suggested-size group 211
keep-hot-count group 20
check-files group yes
Keep in mind that the nscd program caches passwd entries obtained from
This means that when an user is modified on the ldap server, the nscd
cache remains valid. This is avoided when using flat unix files by the
check-files directive that invalidates the cache when the corresponding file is
modified. Such a mechanism should be generalized, at the moment anyway does not
apply to LDAP. A way to avoid possible misalignments between the LDAP server
and the cache is to invalidate the cache manually when updating passwd entries
with the command:
Where TABLE can be passwd, groups or hosts.
To avoid confusion when testing, do not use nscd.
Moreover using nss and nscd will produce a lot of open filedescriptors,
so is easy to run out of available filedescriptors on the system (this can hang
You can increase the maximum number of filedescriptors in a Linux box
(Kernel 2.2.x) with something like:
#echo 16384 > /proc/sys/fs/file-max
The maximum number of filedescriptors suggested for a system depends
anyway from the configuration of your system.
The LDAP client configuration file /etc/ldap.conf is
read by pam_ldap and nss_ldap as well as other LDAP clients. The following is an
example of how it should look like in our environment.
# $Id: section-pamnss.sgml,v 1.2 2001/03/26 16:57:07 rolek Exp $
# This is the configuration file for the LDAP nameservice
# switch library and the LDAP PAM module.
# PADL Software
# If the host and base aren't here, then the DNS RR
# _ldap._tcp.[defaultdomain]. will be resolved. [defaultdomain]
# will be mapped to a distinguished name and the target host
# will be used as the server.
# Your LDAP server. Must be resolvable without using LDAP.
# The distinguished name of the search base.
base dc=yourorg, dc=com
# The LDAP version to use (defaults to 2,
# use 3 if you are using OpenLDAP 2.0.x or Netscape Directory Server)
# ldap_version 3
# The distinguished name to bind to the server with.
# Optional: default is to bind anonymously.
# binddn cn=manager,dc=padl,dc=com
# The credentials to bind with.
# Optional: default is no credential.
# The port.
# Optional: default is 389. 636 is for ldaps
# The search scope.
# The following options are specific to nss_ldap.
# The hashing algorithm your libc uses.
# Optional: default is des
# The following options are specific to pam_ldap.
# Filter to AND with uid=\%s
# The user ID attribute (defaults to uid)
# Search the root DSE for the password policy (works
# with Netscape Directory Server)
# Group to enforce membership of
# Group member attribute
# Template login attribute, default template user
# (can be overridden by value of former attribute
# in user's entry)
# Hash password locally; required for University of
# Michigan LDAP server, and works with Netscape
# Directory Server if you're using the UNIX-Crypt
# hash mechanism and not using the NT Synchronization
# SSL Configuration
Note: To avoid problems with the various applications that may read this file
it is suggested not to use tabs between parameters and values, only a single
The pam_groupdn directive is useful when an LDAP server provides
authentication information to a pool of clients, but the user should be
authorized only on a set of clients. This directive can provide the same
functionality of NIS netgroups.
The SSL configuration directives are not documented in the package, but
they tell to enable SSL and where the file containing the LDAP server
certificate and the CA certificate is stored.
A Netscape certificate database named cert7.db is searched in sslpath.
This file must contain the server certificate and the CA certificate (unless
the server certificate is self signed). There are two ways to generate this
file: using the Netscape PKCS#11 tools or using the Netscape browser.
With the Netscape browser, after you have started slapd and stunnel on the
server you can use Netscape Navigator to connect to the URL
https://your.ldap.server:636/, you will be prompted to insert the server
certificate in your database. Also the CA certificate (provided by your CA) must
be loaded in the database (unless you are using a self signed certificate). At
this point you can copy the $HOME/.netscape/cert7.db in
sslpath. It is preferred that you use a scratch account with
a default cert7.db file since other server certificates,
that may be present in your personal certificate database, will be considered by
your LDAP client as trusted authentication servers. Once the browser has imported
the server certificate it can be used to debug SSL since it will behave like the
pam and nss libraries.
At this point account creation and maintenance should be done using LDAP
Unfortunately these general purpose tools are not intended for Un*x
accounts maintenance. The one that seems to be enough versatile is the LDAP
that allows to set passwords in various formats and can use SSL to connect to the
As it is for NIS with a single master server (no slave servers), LDAP
without a replication mechanism represents a single point of failure for the
authentication system. For authentication purposes it is rather important to
implement LDAP replication. The server that cames with OpenLDAP (slapd) provides