| Linux System Administration |
| Home | Tech | Linux | Links | Consulting |
|
OpenLDAP consumer with MIT Kerberos V on Debian squeeze
This page describes how to set up an OpenLDAP consumer server (syncrepl) with MIT Kerberos V support for authentication and encryption. It depends on a previously installed OpenLDAP provider server, ldapks1.example.com, which also includes Kerberos support, as well as two Kerberos servers, kdc1.example.com and kdc2.example.com − a master and a slave. At least the master KDC will be necessary for this exercise, but it may be more interesting to use the slave. It is assumed that the reader is familiar with the concepts involved in the installation and configuration of an OpenLDAP consumer server. However, as opposed to that configuration, this OpenLDAP system will only store Unix user account information for authorization; the task of authentication will be performed by Kerberos, with all username and password combinations being stored in a Kerberos database. In this example, OpenLDAP is installed on a host running Debian 6.0 (squeeze). If followed properly, the step-by-step process should produce a consumer server with a copy of the DIT from the provider. But before the actual OpenLDAP installation process can begin, it will first be necessary to install the operating system on a new host called ldapks2.example.com. A DNS server must also be available on the network with a zone file to which forward and reverse mappings can be added for this host. After the initial installation of the operating system, make sure these packages are installed on the system as well: ~# apt-get install ntp ntpdate nmap Afterwards, edit /etc/ntp.conf so that the machine synchronizes to a common NTP server (preferably a local one) and edit /etc/default/ntpdate to use the same host also. Now the installation of the new server can begin. 1. Patching the source Because of a bug in the current version of slapd, 2.4.23-7, the source code must first be patched. Without it, slapd will return errors regarding the olcDbURI attribute (necessary for the chaining configuration) and even refuse to (re)start if it is configured that way. So, prepare the new host for a package build by installing the following packages on the new host, ldaps2: ~# apt-get install dpkg-dev devscripts This will result in 100 new packages being installed. Next, obtain the source: ~# cd /usr/src ~# apt-get source openldap Three files will be downloaded and a directory, ./openldap-2.4.23/, will be created that contains the source code. Soon, seven Debian packages will be built from it, but first the various dependencies for these package must be downloaded and installed: ~# apt-get build-dep openldap 70 new packages will be installed as a result. Next, download and apply the patch: ~# wget -q ftp://ftp.openldap.org/incoming/pierangelo-masarati-2010-04-29-chain.1.patch ~# patch openldap-2.4.23/servers/slapd/back-ldap/chain.c < pierangelo-masarati-2010-04-29-chain.1.patch patching file openldap-2.4.23/servers/slapd/back-ldap/chain.c Hunk #1 succeeded at 1137 (offset -3 lines). Hunk #2 succeeded at 1148 (offset -3 lines). ~# _ Then, enter the source code directory, add a suffix to the Debian version number to distinguish the results of this build, and compile the binary package without signing the .changes file: ~# cd openldap-2.4.23/ ~# dch -l patched 'Including a patch by Pierangelo Masarati, 2010-04-29.' ~# debuild -us -uc This build can take a long time − perhaps over an hour − most of which is due to automated testing routines. Eventually, seven new Debian packages will be created in the parent directory, /usr/src/: ldap-utils_2.4.23-7patched1_i386.deb libldap-2.4-2-dbg_2.4.23-7patched1_i386.deb libldap-2.4-2_2.4.23-7patched1_i386.deb libldap2-dev_2.4.23-7patched1_i386.deb slapd-dbg_2.4.23-7patched1_i386.deb slapd-smbk5pwd_2.4.23-7patched1_i386.deb slapd_2.4.23-7patched1_i386.deb 2. Kerberos client install First, run the following command to test if the MIT Kerberos V server installed previously is available on the network: ~# nmap -sU -sT -p U:88,464,T:464,749 kdc1.example.com Starting Nmap 5.00 ( http://nmap.org ) at 2010-12-20 17:13 CET Interesting ports on kdc1.example.com (192.168.2.36): PORT STATE SERVICE 464/tcp open kpasswd5 749/tcp open kerberos-adm 88/udp open|filtered kerberos-sec 464/udp open|filtered kpasswd5 MAC Address: 08:00:27:AB:13:C1 (Cadmus Computer Systems) Nmap done: 1 IP address (1 host up) scanned in 1.42 seconds ~# _ If not all of the above ports are shown as open, fix that problem first. If they are, continue by installing these three packages:
~# apt-get install krb5-{config,user} libpam-krb5
A total of three packages are installed as a result with no dependencies: krb5-config 2.2 Configuration files for Kerberos Version 5 krb5-user 1.8.3+dfsg-4 Basic programs to authenticate using MIT Kerberos libpam-krb5 4.3-1 PAM module for MIT Kerberos During the installation, krb5-config will require that a few questions be answered: Default Kerberos version 5 realm: EXAMPLE.COM Kerberos servers for your realm: kdc1.example.com kdc2.example.com Administrative server for your Kerberos realm: krb.example.com These settings are saved in the Kerberos realm configuration file /etc/krb5.conf. Mostly, it contains information about the realms of a number of other organizations and options regarding Kerberos 4, all of which is unnecessary in this case. The entire contents could be replaced with:
[libdefaults]
default_realm = EXAMPLE.COM
forwardable = true
proxiable = true
[realms]
EXAMPLE.COM = {
kdc = kdc1.example.com
kdc = kdc2.example.com
admin_server = krb.example.com
}
See this section for a more detailed explanation of this file. Regarding the list of KDCs that are specified here, it is often recommended to use a predetermined set of DNS hostname aliases (CNAME records) to refer to the Kerberos servers on a network. However, it is also possible to omit the KDC entries in /etc/krb5.conf and instead rely on DNS SRV resource records to do the same job. See DNS discovery for MIT Kerberos V for information on how to do that. Use kadmin to create a Kerberos principal for the LDAP service and a matching keytab file by issuing a few commands: ~# kadmin -p admin Authenticating as principal admin with password. Password for admin@EXAMPLE.COM: Lampropeltis kadmin: addprinc -randkey ldap/ldapks2.example.com WARNING: no policy specified for ldap/ldapks2.example.com@EXAMPLE.COM; defaulting to no policy Principal "ldap/ldapks2.example.com@EXAMPLE.COM" created. kadmin: ktadd ldap/ldapks2.example.com Entry for principal ldap/ldapks2.example.com with kvno 2, encryption type AES-256 CTS mode with 96-bit SHA-1 HMAC added to keytab WRFILE:/etc/krb5.keytab. Entry for principal ldap/ldapks2.example.com with kvno 2, encryption type ArcFour with HMAC/md5 added to keytab WRFILE:/etc/krb5.keytab. Entry for principal ldap/ldapks2.example.com with kvno 2, encryption type Triple DES cbc mode with HMAC/sha1 added to keytab WRFILE:/etc/krb5.keytab. Entry for principal ldap/ldapks2.example.com with kvno 2, encryption type DES cbc mode with CRC-32 added to keytab WRFILE:/etc/krb5.keytab. kadmin: q ~# _ The -randkey switch is used to avoid having to use a password. To list the keys in /etc/krb5.keytab, use the klist -ke command. A host (or service) principal and a keytab file should be created for and saved on all of the various client machines that are part of a Kerberos realm. 3. OpenLDAP install Before starting the install process, run the following command to test if the previously installed OpenLDAP provider server is actually available on the network: ~# nmap -p 389 ldapks1.example.com Starting Nmap 5.00 ( http://nmap.org ) at 2010-12-20 17:53 CET Interesting ports on ldapks1.example.com (192.168.2.53): PORT STATE SERVICE 389/tcp open ldap MAC Address: 08:00:27:43:43:4A (Cadmus Computer Systems) Nmap done: 1 IP address (1 host up) scanned in 0.23 seconds ~# _ If TCP port 389 is not open, fix that problem first. Otherwise, install three of the newly compiled packages: ~# dpkg -i /usr/src/slapd_2.4.23-7patched1_i386.deb \ /usr/src/ldap-utils_2.4.23-7patched1_i386.deb \ /usr/src/libldap-2.4-2_2.4.23-7patched1_i386.deb During the install process, an administrator password will be requested for slapd. Use contortrix: Administrator password: contortrix Confirm password: contortrix Run the following command to test if the new OpenLDAP server is actually running: ~# nmap -p 389 localhost Starting Nmap 5.00 ( http://nmap.org ) at 2010-12-20 17:53 CET Interesting ports on localhost (127.0.0.1): PORT STATE SERVICE 389/tcp open ldap Nmap done: 1 IP address (1 host up) scanned in 0.23 seconds ~# _ Perform a quick test by generating an LDIF dump of the contents of a the database: ~# slapcat hdb_db_open: database "dc=example,dc=com": unclean shutdown detected; attempting recovery. hdb_db_open: database "dc=example,dc=com": recovery skipped in read-only mode. Run manual recovery if errors are encountered. dn: dc=example,dc=com objectClass: top objectClass: dcObject objectClass: organization o: example.com dc: example structuralObjectClass: organization entryUUID: c868cece-a0ac-102f-8fb8-676e1d316d8b creatorsName: cn=admin,dc=example,dc=com createTimestamp: 20101220174613Z entryCSN: 20101220174613.753133Z#000000#000#000000 modifiersName: cn=admin,dc=example,dc=com modifyTimestamp: 20101220174613Z dn: cn=admin,dc=example,dc=com objectClass: simpleSecurityObject objectClass: organizationalRole cn: admin description: LDAP administrator userPassword:: e1NTSEF9aWtSc29hMERuUkdFY0EvMi9vTEg2MWZZV25RVlFZMnI= structuralObjectClass: organizationalRole entryUUID: c86962f8-a0ac-102f-8fb9-676e1d316d8b creatorsName: cn=admin,dc=example,dc=com createTimestamp: 20101220174613Z entryCSN: 20101220174613.756966Z#000000#000#000000 modifiersName: cn=admin,dc=example,dc=com modifyTimestamp: 20101220174613Z ~# _ Edit /etc/ldap/ldap.conf and use these two lines: BASE dc=example,dc=com URI ldap://ldapks2.example.com/ This configuration file is used to set system-wide defaults for LDAP clients. 4. Syncprov Switch over to the provider server, ldapks1.example.com to make some changes to the LDAP configuration database. Two existing entries will be modified and one new one added. Their current status/absence is as follows:
~# ldapsearch -LLLQY EXTERNAL -H ldapi:/// -b cn=config \
"(|(olcDatabase={1}hdb)(cn=module{0})(olcOverlay={0}syncprov))"
dn: cn=module{0},cn=config
objectClass: olcModuleList
cn: module{0}
olcModulePath: /usr/lib/ldap
olcModuleLoad: {0}back_hdb
dn: olcDatabase={1}hdb,cn=config
objectClass: olcDatabaseConfig
objectClass: olcHdbConfig
olcDatabase: {1}hdb
olcDbDirectory: /var/lib/ldap
olcSuffix: dc=example,dc=com
olcAccess: {0}to attrs=userPassword,shadowLastChange by * none
olcAccess: {1}to attrs=loginShell by self write by users read by * none
olcAccess: {2}to dn.base="" by * read
olcAccess: {3}to * by users read by * none
olcLastMod: TRUE
olcRootDN: uid=admin,ou=people,dc=example,dc=com
olcDbCheckpoint: 512 30
olcDbConfig: {0}set_cachesize 0 2097152 0
olcDbConfig: {1}set_lk_max_objects 1500
olcDbConfig: {2}set_lk_max_locks 1500
olcDbConfig: {3}set_lk_max_lockers 1500
olcDbIndex: objectClass eq
olcDbIndex: uid eq
olcDbIndex: cn eq
olcDbIndex: ou eq
olcDbIndex: dc eq
~# _
Here is a description of the changes that will be made to the cn=config DIT:
To make both of the above changes to the cn=config DIT, first create an LDIF file, called ~/olc-mod3.ldif, with the following contents:
# 1.1.
dn: olcDatabase={1}hdb,cn=config
changetype: modify
add: olcDbIndex
olcDbIndex: entryUUID eq
-
# 1.2.
add: olcDbIndex
olcDbIndex: entryCSN eq
# 2.
dn: cn=module{0},cn=config
changetype: modify
add: olcModuleLoad
olcModuleLoad: syncprov
# 3.
dn: olcOverlay=syncprov,olcDatabase={1}hdb,cn=config
changetype: add
objectClass: olcOverlayConfig
objectClass: olcSyncProvConfig
olcOverlay: {0}syncprov
olcSpCheckpoint: 100 10
olcSpSessionlog: 100
See this section for an explanation of the last two modifications. Apply the changes in the LDIF file with this command:
~# ldapmodify -QY EXTERNAL -H ldapi:/// -f ~/olc-mod3.ldif
modifying entry "olcDatabase={1}hdb,cn=config"
modifying entry "cn=module{0},cn=config"
adding new entry "olcOverlay=syncprov,olcDatabase={1}hdb,cn=config"
~# _
Rerun the previous ldapsearch command to verify that all of the changes have been made successfully. Note that an index number has been added automatically to the RDN for the syncprov overlay. 5. Kstart Return to ldapks2. To keep its copy of the database in sync, the consumer must contact the provider and authenticate itself just like any other client. It has previously been shown how to do this with a simple bind and a password, but in this case Kerberos will be used to take care of both slapd consumer authentication to the provider and encryption of the entire replication process. To make this possible, an initial Kerberos ticket (a TGT) must be obtained automatically on behalf of the openldap user, which owns the slapd process, and it must be renewed regularly. One way to do this is to run kinit -k with a cron job, but a better solution is to use a modified version of kinit, called k5start. Install it with: ~# apt-get install kstart This is the only package that gets installed as a result: kstart 3.16-3 Kerberos kinit supporting AFS and ticket refreshing To configure it, just add this line to the end of the /etc/inittab file to start running k5start in the background soon after the system boots up: KS:2345:respawn:/usr/bin/k5start -U -f /etc/krb5.keytab -K 10 -l 24h -k /tmp/krb5cc_105 -o openldap A number of options have been used for this command:
After saving this modification to /etc/inittab, start k5start for the first time by forcing init to reload its configuration file: ~# kill -HUP 1 A new ticket cache file, /tmp/krb5cc_105, should be created almost immediately as a consequence. 6. Slapd kerberization This step is about kerberizing OpenLDAP. A number of requirements have already been met, but there is more. First, install this one package: ~# apt-get install libsasl2-modules-gssapi-mit Only one package is installed as a result with no dependencies: libsasl2-modules-gssapi-mit 2.1.23.dfsg1-6 Cyrus SASL - pluggable authentication modules (GSSAPI) First, change the permissions and ownership of the Kerberos service keytab file to allow slapd to read it: ~# chmod 640 /etc/krb5.keytab ~# chown root.openldap /etc/krb5.keytab ~# _ Next, three different files must be modified. First, edit /etc/default/slapd and uncomment a line near the end that exports as a variable the location of the Kerberos system keytab file: export KRB5_KTNAME=/etc/krb5.keytab Edit /etc/ldap/ldap.conf and add the following line, which specifies the authentication mechanism, to the end of the file: SASL_MECH GSSAPI The last part of this kerberization step is, just as on ldapks1, to get slapd to map the GSSAPI-format user names to LDAP names. To do this, only the root object of the cn=config DIT needs to be modified. Currently, it looks like this: ~# ldapsearch -LLLQY EXTERNAL -H ldapi:/// -b cn=config cn=config dn: cn=config objectClass: olcGlobal cn: config olcArgsFile: /var/run/slapd/slapd.args olcLogLevel: none olcPidFile: /var/run/slapd/slapd.pid olcToolThreads: 1 ~# _ Here is a description of the changes that will be made to ldaps2's runtime configuration:
To make the above changes to the slapd runtime configuration on ldapks2, create an LDIF file, called ~/olc-mod1.ldif, with the following contents: # 1.1. dn: cn=config changetype: modify replace: olcLogLevel olcLogLevel: stats - # 1.2. add: olcAuthzRegexp olcAuthzRegexp: uid=([^,]+),cn=example.com,cn=gssapi,cn=auth uid=$1,ou=people,dc=example,dc=com - # 1.3. add: olcSaslRealm olcSaslRealm: EXAMPLE.COM Th authz-regexp statement used here is exactly the same as on the provider server, ldapks1. Apply the changes in the LDIF file with this command: ~# ldapmodify -QY EXTERNAL -H ldapi:/// -f ~/olc-mod1.ldif modifying entry "cn=config" ~# _ Rerun the previous ldapsearch command to verify that all of the changes have been made successfully. Because of the new software that has been added and the changes that have also been made outside of the slapd server, it must be restarted before they will all have the desired effect: ~# /etc/init.d/slapd restart Stopping OpenLDAP: slapd. Starting OpenLDAP: slapd. ~# _ 7. Syncrepl The next set of changes to be made to the slapd runtime configuration on ldaps2 will enable the replication process for this system as a slapd consumer. For that, only the database definition object needs to be changed, which currently looks like this:
~# ldapsearch -LLLQY EXTERNAL -H ldapi:/// -b cn=config olcDatabase={1}hdb
dn: olcDatabase={1}hdb,cn=config
objectClass: olcDatabaseConfig
objectClass: olcHdbConfig
olcDatabase: {1}hdb
olcDbDirectory: /var/lib/ldap
olcSuffix: dc=example,dc=com
olcAccess: {0}to attrs=userPassword,shadowLastChange by self write by anonymou
s auth by dn="cn=admin,dc=example,dc=com" write by * none
olcAccess: {1}to dn.base="" by * read
olcAccess: {2}to * by self write by dn="cn=admin,dc=example,dc=com" write by *
read
olcLastMod: TRUE
olcRootDN: cn=admin,dc=example,dc=com
olcRootPW: {SSHA}ikRsoa0DnRGEcA/2/oLH61fYWnQVQY2r
olcDbCheckpoint: 512 30
olcDbConfig: {0}set_cachesize 0 2097152 0
olcDbConfig: {1}set_lk_max_objects 1500
olcDbConfig: {2}set_lk_max_locks 1500
olcDbConfig: {3}set_lk_max_lockers 1500
olcDbIndex: objectClass eq
~# _
Here is a description of the changes that will be made to the database definition object on ldaps2:
To make the above changes to the slapd runtime configuration on ldapks2, create an LDIF file, called ~/olc-mod2.ldif, with the following contents:
# 1.1.1.
dn: olcDatabase={1}hdb,cn=config
changetype: modify
delete: olcAccess
olcAccess: {2}to *
by self write
by dn="cn=admin,dc=example,dc=com" write
by * read
-
# 1.1.2.
delete: olcAccess
olcAccess: {1}to dn.base=""
by * read
-
# 1.1.3.
delete: olcAccess
olcAccess: {0}to attrs=userPassword,shadowLastChange
by self write
by anonymous auth
by dn="cn=admin,dc=example,dc=com" write
by * none
-
# 1.2.1.
add: olcAccess
olcAccess: {0}to attrs=userPassword,shadowLastChange
by * none
-
# 1.2.2.
add: olcAccess
olcAccess: {2}to dn.base=""
by * read
-
# 1.2.3.
add: olcAccess
olcAccess: {3}to *
by users read
by * none
-
# 1.3.
replace: olcRootDN
olcRootDN: cn=manager
-
# 1.4.
delete: olcRootPW
-
# 1.5.
add: olcDbIndex
olcDbIndex: entryCSN eq
-
# 1.6.
add: olcDbIndex
olcDbIndex: entryUUID eq
-
# 1.7.
add: olcDbIndex
olcDbIndex: uid eq
-
# 1.8.
add: olcDbIndex
olcDbIndex: cn eq
-
# 1.9.
add: olcDbIndex
olcDbIndex: ou eq
-
# 1.10.
add: olcDbIndex
olcDbIndex: dc eq
-
# 1.11.
add: olcSyncrepl
olcSyncrepl: rid=123
provider="ldap://ldapks.example.com:389/"
type=refreshAndPersist
retry="60 30 300 +"
searchbase="dc=example,dc=com"
bindmethod=sasl
saslmech=gssapi
The presence of the olcSyncrepl attribute will indicate that the current database is a replica. This directive has many different options, a number of which are used here:
The above syncrepl configuration uses persistent push-based synchronization in which the provider keeps track of the consumer, sending it the necessary updates whenever its own database is modified. This is often desirable when modifications are frequent, but if not, use the refreshOnly operation type instead: # 2.11. (alternative) add: olcSyncrepl olcSyncrepl: rid=123 provider="ldap://ldapks.example.com:389/" type=refreshOnly interval=00:00:05:00 searchbase="dc=example,dc=com" bindmethod=sasl saslmech=gssapi
This alternative configuration features the polling, pull-based refreshOnly operation type. The interval setting used here, 00:00:05:00, means that replication will occur once every five minutes. Anyway, apply the changes in the LDIF file with this command:
~# ldapmodify -QY EXTERNAL -H ldapi:/// -f ~/olc-mod2.ldif
modifying entry "olcDatabase={1}hdb,cn=config"
~# _
Rerun the previous ldapsearch command to verify that all of the changes have been made successfully. At this point the consumer will already have synchronized its database with the provider. A quick slapcat will confirm this. However, the two copies of the database are not really in sync: the root object is still the same and the default cn=admin object is still present. To fix this situation, the consumer's database must resynchronized with the provider: ~# /etc/init.d/slapd stop Stopping OpenLDAP: slapd. ~# rm /var/lib/ldap/* ~# /etc/init.d/slapd start Starting OpenLDAP: slapd. ~# _ This will stop the LDAP service on ldapks2, delete its copy of the database and restart the service, at which point it will fetch a fresh copy from the provider. Run slapcat once more to see the difference. 8. Referrals LDAP Sync Replication does present one important problem for LDAP clients: the database containing the DIT on a consumer server is always read-only. To allow clients to make modifications to the DIT anyway, the consumer must either refer its clients on to the provider, or proxy their requests. Both of these options will be arranged by modifying two existing entries in the consumer server's configuration DIT and adding two new ones. The following search command on ldapks2 will verify their current status/absence:
~# ldapsearch -LLLQY EXTERNAL -H ldapi:/// -b cn=config "(|(cn=module{0})\
(olcDatabase={1}hdb)(olcOverlay={0}chain)(olcDatabase={0}ldap))"
dn: cn=module{0},cn=config
objectClass: olcModuleList
cn: module{0}
olcModulePath: /usr/lib/ldap
olcModuleLoad: {0}back_hdb
dn: olcDatabase={1}hdb,cn=config
objectClass: olcDatabaseConfig
objectClass: olcHdbConfig
olcDatabase: {1}hdb
olcDbDirectory: /var/lib/ldap
olcSuffix: dc=example,dc=com
olcAccess: {0}to attrs=userPassword,shadowLastChange by * none
olcAccess: {1}to dn.base="" by * read
olcAccess: {2}to * by users read by * none
olcLastMod: TRUE
olcRootDN: cn=manager
olcSyncrepl: {0}rid=123 provider="ldap://ldapks.example.com:389/" type=refresh
AndPersist retry="60 30 300 +" searchbase="dc=example,dc=com" bindmethod=sasl
saslmech=gssapi
olcDbCheckpoint: 512 30
olcDbConfig: {0}set_cachesize 0 2097152 0
olcDbConfig: {1}set_lk_max_objects 1500
olcDbConfig: {2}set_lk_max_locks 1500
olcDbConfig: {3}set_lk_max_lockers 1500
olcDbIndex: objectClass eq
olcDbIndex: entryCSN eq
olcDbIndex: entryUUID eq
olcDbIndex: uid eq
olcDbIndex: cn eq
olcDbIndex: ou eq
olcDbIndex: dc eq
~# _
Here is a description of the changes that will be made to the cn=config DIT on ldapks2:
To make the above changes to the cn=config DIT on ldapks2, create an LDIF file, called ~/olc-mod3.ldif, with the following contents:
# 1.
dn: olcDatabase={1}hdb,cn=config
changetype: modify
add: olcUpdateref
olcUpdateref: "ldap://ldapks.example.com:389/"
# 2.
dn: cn=module{0},cn=config
changetype: modify
add: olcModuleLoad
olcModuleLoad: {1}back_ldap
# 3.
dn: olcOverlay=chain,olcDatabase={-1}frontend,cn=config
changetype: add
objectClass: olcOverlayConfig
objectClass: olcChainConfig
olcOverlay: {0}chain
olcChainReturnError: TRUE
# 4.
dn: olcDatabase=ldap,olcOverlay={0}chain,olcDatabase={-1}frontend,
cn=config
changetype: add
objectClass: olcLDAPConfig
objectClass: olcChainDatabase
olcDatabase: {0}ldap
olcDbURI: "ldap://ldapks.example.com:389/"
olcDbRebindAsUser: TRUE
olcDbIDAssertBind: bindmethod=sasl
saslmech=gssapi
mode=self
Here, the olcDbIDAssertBind directive defines the parameters of the authentication method used by ldapks2 to authorize connections from authenticated users. It has a number of different options, a few of which are used here:
Apply the above changes on ldapks2 with this command:
~# ldapmodify -QY EXTERNAL -H ldapi:/// -f ~/olc-mod3.ldif
modifying entry "olcDatabase={1}hdb,cn=config"
modifying entry "cn=module{0},cn=config"
adding new entry "olcOverlay=chain,olcDatabase={1}hdb,cn=config"
adding new entry "olcDatabase=ldap,olcOverlay={0}chain,olcDatabase={1}hdb,cn=config"
~# _
Rerun the previous ldapsearch command to verify that all of the changes have been made successfully. 9. Authentication test Run some tests on ldapks2. First try a simple unauthenticated (-x) LDAP query: ~# ldapsearch -x -LLL No such object (32) ~# _ This should not work, because modification 1.2.4 in step 9 prevents unauthenticated users from reading the database. However, the authenticated version of this query will currently give an error: ~# ldapsearch -LLL SASL/GSSAPI authentication started ldap_sasl_interactive_bind_s: Local error (-2) additional info: SASL(-1): generic failure: GSSAPI Error: Unspecified GSS failure. Minor code may provide more information (Credentials cache file '/tmp/krb5cc_0' not found) ~# _ The solution is to first acquire a Kerberos ticket for the admin user (password Lampropeltis): ~# kinit admin Password for admin@EXAMPLE.COM: Lampropeltis ~# _ A verification of the ticket should show a success: ~# klist Ticket cache: FILE:/tmp/krb5cc_0 Default principal: admin@EXAMPLE.COM Valid starting Expires Service principal 12/22/10 21:08:32 12/23/10 21:08:26 krbtgt/EXAMPLE.COM@EXAMPLE.COM 12/22/10 21:08:37 12/23/10 21:08:26 ldap/ldapks2.example.com@EXAMPLE.COM ~# _ Now the authenticated version of the query should work: ~# ldapsearch -LLL SASL/GSSAPI authentication started SASL username: admin@EXAMPLE.COM SASL SSF: 56 SASL data security layer installed. dn: dc=example,dc=com objectClass: top objectClass: dcObject objectClass: organization o: example.com dc: example dn: ou=people,dc=example,dc=com objectClass: organizationalUnit ou: people dn: ou=groups,dc=example,dc=com objectClass: organizationalUnit ou: groups ~# _ 10. Proxy authorization Before the chaining configuration added in step 10 will work, more changes first have to be made to the OpenLDAP provider server, ldapks1. What is required is something called SASL proxy authorization. With this mechanism it is possible to allow an authenticated user, such as the consumer server in this example, to assume the identity of other users and use that to relay their write requests on to the provider server. The ability to assume another user's identity is important, since it would otherwise not be possible to enforce existing access rules. Two changes are needed to enable proxy authorization. First, a new LDAP object must be created that corresponds to the consumer user's GSSAPI (Kerberos) identity and includes an authzTo attribute that will allow it to act as an authorization proxy for certain users. A new organizational unit will also be created in which the new consumer server object will reside. Second, the provider server, ldapks1, will be configured to accept such proxy requests. The new LDAP object to represent ldapks2 is required because of the need to authorize it to act as a proxy entity. At the moment it is only able to replicate with ldapks1 thanks to Kerberos authentication, but without an LDAP object to represent it as well, any form of LDAP authorization will not be possible. An examination of /var/log/syslog on ldapks1 will show that, when ldapks2 binds for the purpose of replication, its LDAP identity is currently part of ou=people:
BIND authcid="ldap/ldapks2.example.com" \
authzid="ldap/ldapks2.example.com"
BIND dn="uid=ldap/ldapks2.example.com,ou=people,dc=example,dc=com" \
mech=GSSAPI sasl_ssf=56 ssf=56
This must be changed. On ldapks1, start by creating an LDIF file, called ~/ldapks2.ldif, with the following contents:
dn: ou=consumers,dc=example,dc=com
objectClass: organizationalUnit
ou: consumers
dn: cn=ldapks2,ou=consumers,dc=example,dc=com
cn: ldapks2
objectClass: simpleSecurityObject
objectClass: organizationalRole
description: LDAP server2 replicator
authzTo: dn.regex:^uid=[^,]+,ou=people,dc=example,dc=com$
userPassword: {CRYPT}*
This will create and origanizational unit and an organizationalRole object. Note the last two attributes: authzTo is a source rule that determines which user identities it is allowed to assume; in this case its value is a regular expression that matches all uid objects in the ou=people,dc=example,dc=com container. The userPassword attribute, on the other hand, is only included because it is required for organizationalRole objects, but since it is not actually necessary here, an invalid hash is used for its value. To add these new objects, authenticate as the administrative user and add them to the database using the following commands: root@ldapks1:~# kinit admin Password for admin@EXAMPLE.COM: Lampropeltis root@ldapks1:~# ldapadd -f ~/ldapks2.ldif SASL/GSSAPI authentication started SASL username: admin@EXAMPLE.COM SASL SSF: 56 SASL data security layer installed. adding new entry "ou=consumers,dc=example,dc=com" adding new entry "cn=ldapks2,ou=consumers,dc=example,dc=com" root@ldapks1:~# _ Now to make the second of the two changes. To ensure that the provider server, ldapks1, will accept proxy requests associated with the authzTo attribute, its cn=config DIT must be modified. This involves two entries that currently look like this:
~# ldapsearch -LLLQY EXTERNAL -H ldapi:/// -b cn=config \
"(|(cn=config)(olcDatabase={1}hdb))"
dn: cn=config
objectClass: olcGlobal
cn: config
olcArgsFile: /var/run/slapd/slapd.args
olcAuthzRegexp: {0}uid=([^,]+),cn=example.com,cn=gssapi,cn=auth uid=$1,ou=peop
le,dc=example,dc=com
olcLogLevel: stats
olcPidFile: /var/run/slapd/slapd.pid
olcSaslRealm: EXAMPLE.COM
olcToolThreads: 1
dn: olcDatabase={1}hdb,cn=config
objectClass: olcDatabaseConfig
objectClass: olcHdbConfig
olcDatabase: {1}hdb
olcDbDirectory: /var/lib/ldap
olcSuffix: dc=example,dc=com
olcAccess: {0}to attrs=userPassword,shadowLastChange by * none
olcAccess: {1}to attrs=loginShell by self write by users read by * none
olcAccess: {2}to dn.base="" by * read
olcAccess: {3}to * by users read by * none
olcLastMod: TRUE
olcRootDN: uid=admin,ou=people,dc=example,dc=com
olcDbCheckpoint: 512 30
olcDbConfig: {0}set_cachesize 0 2097152 0
olcDbConfig: {1}set_lk_max_objects 1500
olcDbConfig: {2}set_lk_max_locks 1500
olcDbConfig: {3}set_lk_max_lockers 1500
olcDbIndex: objectClass eq
olcDbIndex: uid eq
olcDbIndex: cn eq
olcDbIndex: ou eq
olcDbIndex: dc eq
olcDbIndex: entryUUID eq
olcDbIndex: entryCSN eq
~# _
Here is a description of the changes that will be made to the cn=config tree on ldapks1 to enable proxy authorization:
To make the above changes in the same order to the cn=config DIT, create an LDIF file on ldapks1, called ~/olc-mod4.ldif, with the following contents:
# 1.1.1.
dn: cn=config
changetype: modify
delete: olcAuthzRegexp
olcAuthzRegexp: {0}uid=([^,]+),cn=example.com,cn=gssapi,cn=auth
uid=$1,ou=people,dc=example,dc=com
-
# 1.1.2.
add: olcAuthzRegexp
olcAuthzRegexp: {0}uid=ldap/([^/\.]+).example.com,cn=example.com,
cn=gssapi,cn=auth cn=$1,ou=consumers,dc=example,dc=com
-
# 1.1.3.
add: olcAuthzRegexp
olcAuthzRegexp: {1}uid=([^,]+),cn=example.com,cn=gssapi,cn=auth
uid=$1,ou=people,dc=example,dc=com
-
# 1.2.
add: olcAuthzPolicy
olcAuthzPolicy: to
# 2.1.1.
dn: olcDatabase={1}hdb,cn=config
changetype: modify
delete: olcAccess
olcAccess: {0}to attrs=userPassword,shadowLastChange
by * none
-
# 2.1.2.
add: olcAccess
olcAccess: {0}to attrs=userPassword,shadowLastChange
by dn.one="ou=consumers,dc=example,dc=com" read
by * none
Run this command on ldapks1 to apply the change:
~# ldapmodify -QY EXTERNAL -H ldapi:/// -f ~/olc-mod4.ldif
modifying entry "cn=config"
modifying entry "olcDatabase={1}hdb,cn=config"
~# _
Rerun the previous search command to verify that the changes have been made correctly. For some reason, in this particular situation, despite the runtime configuration, slapd still has to be restarted before proxy authorization will work: ~# /etc/init.d/slapd restart Starting OpenLDAP: slapd. ~# _ 11. Proxy auth test This last step will test whether the newly added proxy authorization actually works. This is easily done on the consumer server, ldapks2, as the admin user, using the ldapadd command. If it works, the newly added entries will become immediately available on the consumer server (assuming refreshAndPersist replication is used). First, on ldapks2, create a file, called ~/bbeamon.ldif to add a new user account based on an existing Kerberos account by the same name, and add this information to it:
dn: cn=bbeamon,ou=groups,dc=example,dc=com
cn: bbeamon
gidNumber: 20002
objectClass: top
objectClass: posixGroup
dn: uid=bbeamon,ou=people,dc=example,dc=com
uid: bbeamon
uidNumber: 20002
gidNumber: 20002
cn: Bob
sn: Beamon
objectClass: top
objectClass: person
objectClass: posixAccount
objectClass: shadowAccount
loginShell: /bin/bash
homeDirectory: /home/bbeamon
userPassword: {CRYPT}*
Authenticate first if necessary: ~# kinit admin Password for admin@EXAMPLE.COM: Lampropeltis ~# _ Then add the new user: ~# ldapadd -f bbeamon.ldif SASL/GSSAPI authentication started SASL username: admin@EXAMPLE.COM SASL SSF: 56 SASL data security layer installed. adding new entry "cn=bbeamon,ou=groups,dc=example,dc=com" adding new entry "uid=bbeamon,ou=people,dc=example,dc=com" ~# _ If that works, the following entries should be seen in the /var/log/syslog file on ldapks1 as evidence of the proxy event that just transpired: conn=1000 fd=15 ACCEPT from IP=192.168.2.56:41914 (IP=0.0.0.0:389) conn=1000 op=0 BIND dn="" method=163 conn=1000 op=0 RESULT tag=97 err=14 text=SASL(0): successful result: conn=1000 op=1 BIND dn="" method=163 conn=1000 op=1 RESULT tag=97 err=14 text=SASL(0): successful result: conn=1000 op=2 BIND dn="" method=163 conn=1000 op=2 BIND authcid="ldap/ldapks2.example.com@EXAMPLE.COM" authzid="ldap/ldapks2.example.com@EXAMPLE.COM" conn=1000 op=2 BIND dn="cn=ldapks2,ou=consumers,dc=example,dc=com" mech=GSSAPI sasl_ssf=56 ssf=56 conn=1000 op=2 RESULT tag=97 err=0 text= connection_input: conn=1000 deferring operation: binding conn=1000 op=3 PROXYAUTHZ dn="uid=admin,ou=people,dc=example,dc=com" conn=1000 op=3 ADD dn="cn=bbeamon,ou=groups,dc=example,dc=com" conn=1000 op=3 RESULT tag=105 err=0 text= conn=1001 fd=22 ACCEPT from IP=192.168.2.56:41915 (IP=0.0.0.0:389) conn=1001 op=0 BIND dn="" method=163 conn=1001 op=0 RESULT tag=97 err=14 text=SASL(0): successful result: conn=1001 op=1 BIND dn="" method=163 conn=1001 op=1 RESULT tag=97 err=14 text=SASL(0): successful result: conn=1001 op=2 BIND dn="" method=163 conn=1001 op=2 BIND authcid="ldap/ldapks2.example.com@EXAMPLE.COM" authzid="ldap/ldapks2.example.com@EXAMPLE.COM" conn=1001 op=2 BIND dn="cn=ldapks2,ou=consumers,dc=example,dc=com" mech=GSSAPI sasl_ssf=56 ssf=56 conn=1001 op=2 RESULT tag=97 err=0 text= connection_input: conn=1001 deferring operation: binding conn=1001 op=3 PROXYAUTHZ dn="uid=admin,ou=people,dc=example,dc=com" conn=1001 op=3 ADD dn="uid=bbeamon,ou=people,dc=example,dc=com" conn=1001 op=3 RESULT tag=105 err=0 text= Further examination of the syslog on ldapks1 should also show that whenever ldapks2 is restarted, its replication process binds to the provider with the new LDAP name created for it in step 12: conn=1003 fd=15 ACCEPT from IP=192.168.2.56:34012 (IP=0.0.0.0:389) conn=1003 op=0 BIND dn="" method=163 conn=1003 op=0 RESULT tag=97 err=14 text=SASL(0): successful result: conn=1003 op=1 BIND dn="" method=163 conn=1003 op=1 RESULT tag=97 err=14 text=SASL(0): successful result: conn=1003 op=2 BIND dn="" method=163 conn=1003 op=2 BIND authcid="ldap/ldapks2.example.com@EXAMPLE.COM" authzid="ldap/ldapks2.example.com@EXAMPLE.COM" conn=1003 op=2 BIND dn="cn=ldapks2,ou=consumers,dc=example,dc=com" mech=GSSAPI sasl_ssf=56 ssf=56 conn=1003 op=2 RESULT tag=97 err=0 text= conn=1003 op=3 SRCH base="dc=example,dc=com" scope=2 deref=0 filter="(objectClass=*)" conn=1003 op=3 SRCH attr=* + 12. Bug warning Unfortunately, there is another bug in the current version of slapd for which there is, as yet, no patch. It only affects proxy authorization when using SASL binds with the GSSAPI mechanism − replication is unaffected. With version 2.4.23-7, it may be possible to install the provider and a consumer and complete the above test successfully, but this good behavior is likely to cease later on, perhaps after restarting one or both systems. The problem is that, when the consumer authenticates to the provider, it inexplicably does so with a SIMPLE bind (method=128), instead of with SASL-GSSAPI (method=163). Consequently, no new objects can be added to the DIT from the consumer and the client returns with an error: ~# ldapadd -Qf bbeamon.ldif adding new entry "cn=bbeamon,ou=groups,dc=example,dc=com" ldap_add: Strong(er) authentication required (8) ~# _ Hopefully, a fix will become available soon. When it does, the instructions for applying it will be included here. 13. See also
14. Further reading
15. Sources
Last modified: 2018-03-10, 18:43
©2003-2020 RJ Systems. Permission is granted to copy, distribute and/or modify the content of this page under the terms of the OpenContent License, version 1.0. |
