Netscape Directory Server Installation Guide: Using Custom Install

The setup program will now attempt to bind to your supplier server using the replication DN and password that you supplied. If the supplier server is not running, or if you have not yet created the replication DN on the supplier server, setup will indicate that it could not bind to the supplier server. However, the setup script will continue. Once your consumer server has been installed, you should go to your supplier server and create the directory entry and access control permissions that the consumer requires for replication purposes.

For replication synch interval, enter how long you want the consumer server to wait before checking to see if any updates are required. A value of 0 causes the consumer server to always stay in synch (there is no interval between update attempts).

For replication days, specify the days of the week on which replication can begin. Values are an integer, 0-6, representing a day of the week. That is, 0=Sunday, 1=Monday, 2=Tuesday, and so forth. To specify multiple days of the week, enter each integer with no delimiter between them. Thus, to replicate on Sunday and Wednesday, enter 03. To replicate every day of the week, select the default (all).

For replication time, enter the range of time during which replication can begin. Specify times in 24 hour format. For example, 0100-2300 allows replication to begin anytime between 1 a.m. and 11 p.m. To allow replication to begin any time during the day, select the default (all day).

For disable schema checking, select the default unless you think you may have a problem with old schema. It is strongly recommended that you clean up your directory schema before you import your LDIF to the Directory Server. Further, you should NOT run a production Directory Server with schema checking turned off.

For administration port number, enter a value that is not in use. Be sure to record this value.

For the IP address to bind to, select the default unless you want the Administration Server to listen to some other IP address (this may be necessary for some multi-homed systems). If so, then enter that other IP address.

For Server Administrator ID, enter the user ID and password that you will use to authenticate to the Administration Server when the Directory Server is not running. For best results, this user ID/password pair should be the same as the user ID/password that you specified for the Configuration Directory Administrator.

For the user you want to run the Administration Server as, enter Root. This is the default. For information on why you should run the Administration Server as root, see "Decide Which User and Group to Run Netscape Servers As (Unix only)".

CIR Consumer Server Installs on Windows NT

If you have not already done so, download the product binaries file to the to the machine on which you want to install the server.

Unzip the product binaries files and run the setup program.

When you are asked what you would like to install, select the default, Netscape Servers.

When you are asked what type of installation you would like to perform, select Custom Installation.

For server root, enter a full path to the location where you want to install your server. The location that you enter must be some directory other than the directory from which you are running setup. If the directory that you specify does not exist, setup creates it for you.

For Netscape Server Family Core Components, Netscape Directory Suite, and NT Synchronization Service, select all the components except for the NT Synchronization Service. (NT Synchronization Service installation is described separately in Chapter 5, "Installing and Configuring the Synch Service."

For configuration directory, select "Use existing configuration Directory Server." This allows you to use your existing supplier server as the configuration directory. If you want this consumer server to be its own configuration directory, then select "This instance will be the configuration directory." The next few steps assume that you are using an existing configuration directory. If not then skip to step 11.

For host and port, enter the fully qualified domain name of the machine on which the configuration directory is running, and then enter the port number that the configuration directory is listening to.

For Bind As and password, enter the configuration directory user ID and password. That is, enter the user ID and password that you log in as when you want to authenticate to the console with full privileges. This user ID and password was created when the configuration directory was installed.

For Administration Domain, enter the domain that you want this server to belong to. The name that you enter should be a unique string that is descriptive of the organization responsible for administering the domain. For information on administration domains, see "Determine the Administration Domain".

For a directory suffix, enter a distinguished name meaningful to your enterprise. This string is used to form the name of all your organization's directory entries. Therefore, pick some name that is representative of your organization. It is recommended that you pick a suffix that corresponds to your internet DNS name.

For example, if your organization uses the DNS name airius.com, then enter o=airius.com here.

For server identifier, enter some unique value (normally the default is sufficient). This value is used as part of the name of the directory in which the Directory Server instance is installed. For example, if your machine's host name is phonebook then this name is the default and selecting it will cause the Directory Server instance to be installed into a directory labeled slapd-phonebook.

For the Directory Server port, select the default (389) unless you already have another application using that port.

For Directory Manager DN, enter the distinguished name that you will use when managing the contents of your directory with unlimited privileges (in former releases of the Directory Server, the Directory Manager was known as the root DN). This is the entry that you bind to the directory as when you want access control to be ignored. This distinguished name can be short and does not have to conform to any suffix configured for your directory. It also should not correspond to an actual entry stored in your directory. Examples of possible directory manager DNs are:

cn=Directory Manager

cn=root

uid=admin

For Directory Manager password, enter a value that is at least 8 characters long.

For Configure server for replication, select "A consumer that will pull updates from another server (CIR)".

For host and port, enter the fully qualified DNS host name of the supplier server and the port that it is listening to.

For Bind As and password, enter the distinguished name on the supplier server that has read, search, and compare privileges to the replicated tree, and to the supplier server's changelog directory.

For the Root of Replication, enter the distinguished name of the root point of the directory tree that this consumer server will receive. That is, if you are replicating the tree under ou=people, o=airius.com, then enter that DN here.

For replication days, specify the days of the week on which replication can begin. To replicate every day of the week, select the default (all).

For replication time, enter the range of time during which replication can begin. To allow replication to begin at any time during the day, select the default (all day).

For replication synch interval, enter how long you want the consumer server to wait before checking to see if any updates are required.

For disable schema checking, select the default unless you think you may have a problem with old schema. It is strongly recommended that you clean up your directory schema before you import your LDIF to the Directory Server. Further, you should NOT run a production Directory Server with schema checking turned off.

For administration port number, enter a value that is not in use. Be sure to record this value.

For Administration Server IP Address, select the default unless you want the Administration Server to listen to some other IP address (this may be necessary for some multi-homed systems). If so, then enter that other IP address.

For Administration Server Authentication, enter the user ID and password that you will use to authenticate to the Administration Server when the Directory Server is not running. For best results, this user ID/password pair should be the same as the user ID/password that you use for the Configuration Directory Administrator.

Creating an SIR Supplier Server

This section describes how to create a supplier server for SIR using custom install. The procedure is described first for Unix installations, and then for Windows NT. (See "SIR Supplier Server Installs on Windows NT".) For more information about SIR, refer to the Netscape Directory Server Administrator's Guide.

SIR Supplier Server Installs on Unix

If you have not already done so, download the product binaries file to the to the machine on which you want to install the server.

Unpack the product binaries file using the following command:

# gzip -dc <filename>.tar.gz | tar -xvof-

where <filename> corresponds to the product binaries that you want to unpack.

Run the setup program. You can find it in the directory where you untarred installation files.

When you are asked what you would like to install, select the default, Netscape Servers.

When you are asked what type of installation you would like to perform, select the Custom Installation.

For server root, enter a full path to the location where you want to install your server. The location that you enter must be some directory other than the directory from which you are running setup. If the directory that you specify does not exist, setup creates it for you.

For Netscape Server Family, Netscape Server Family Core Components, Netscape Directory Suite, and Administration Services components, select the default (all components).

For host name, select the default (which is the local host).

For the user and group to run the servers as, enter the identity that you want this server to run as. For more information on the user and groups that you should use when running Netscape servers, see "Decide Which User and Group to Run Netscape Servers As (Unix only)".

For configuration directory, select the default if this directory will host your o=NetscapeRoot tree. Otherwise, enter Yes. You will then be asked for the contact information for the configuration directory. If the server you are currently installing is not the configuration directory, then the configuration directory must exist before you can continue this installation.

For user directory, select the default, No.

For the Directory Server port, select the default (389) unless you already have another application using that port.

For server identifier, enter some unique value (normally the default is sufficient). This value is used as part of the name of the directory in which the Directory Server instance is installed. For example, if your machine's host name is phonebook then this name is the default and selecting it will cause the Directory Server instance to be installed into a directory labeled slapd-phonebook.

For Configuration Directory Administrator ID and password, enter the name and password that you log in as when you want to authenticate to the console with full privileges.

For a directory suffix, enter a distinguished name meaningful to your enterprise. This string is used to form the name of all your organization's directory entries. Therefore, pick some name that is representative of your organization. It is recommended that you pick a suffix that corresponds to your internet DNS name.

For example, if your organization uses the DNS name airius.com, then enter o=airius.com here.

For Directory Manager DN, enter the distinguished name that you will use when managing the contents of your directory with unlimited privileges (in former releases of the Directory Server, the Directory Manager was known as the root DN). This is the entry that you bind to the directory as when you want access control to be ignored. This distinguished name can be short and does not have to conform to any suffix configured for your directory. It also should not correspond to an actual entry stored in your directory. Examples of possible directory manager DNs are:

cn=Directory Manager

cn=root

uid=admin

For Directory Manager password, enter a value that is at least 8 characters long.

For Administration Domain, enter the domain that you want this server to belong to. The name that you enter should be a unique string that is descriptive of the organization responsible for administering the domain. For information on administration domains, see "Determine the Administration Domain".

For set up replication, enter Yes.

For set up a supplier server, select the first option (SIR).

For change log suffix and change log database directory, select the defaults. This is the suffix used by the directory tree that will store all changes made to the supplier server, and the location of the database used to store these changes. These changes must be tracked so that the supplier can know how to update it's consumer servers.

For consumer host name and port, enter the fully qualified DNS name and port of the machine where a consumer server will be running.

For the replication DN and password, enter the supplier DN and password for the consumer server. For example, cn=supplier. By entering this value here, you are only telling the supplier server how to bind to the consumer server; you are not actually setting up the supplier DN on the consumer server. For information on how to set up a supplier DN, see the Netscape Directory Server Administrator's Guide.

For the tree to replicate, enter the distinguished name of the root point of the directory tree that this consumer server will receive. That is, if you are replicating the tree under ou=people, o=airius.com, then enter that DN here.

At this point the setup program will attempt to bind to your consumer server using the DN and password that you supplied. If the consumer server is not running, or if you have not yet created the supplier DN on the consumer server, setup will indicate that it could not bind to the consumer server. However, the setup script will continue. Once your supplier server has been installed, you should go to your consumer server and create the supplier DN required for replication purposes.

For replication days, specify the days of the week on which replication can begin. Values are an integer, 0-6, representing a day of the week. That is, 0=Sunday, 1=Monday, 2=Tuesday, and so forth. To specify multiple days of the week, enter each integer with no delimiter between them. Thus, to replicate on Sunday and Wednesday, enter 03. To replicate every day of the week, select the default (all).

For replication time, enter the range of time during which replication can begin. Specify times in 24 hour format. For example, 0100-2300 allows replication to begin anytime between 1 a.m. and 11 p.m. To allow replication to begin any time during the day, select the default (all day).

Now you are asked if you want to set up this supplier server to also be a consumer server. Do this if you are using replication chaining. Otherwise, select the default (NONE). For information on the questions you will be asked during a consumer server setup, see "SIR Consumer Server Installs on Unix" or "CIR Consumer Server Installs on Unix".

For sample entries, select the default No.

For import an LDIF file, enter the full path to an LDIF file if you have one available. Otherwise, enter the word suggest. Suggest creates a simple directory tree that conforms to Netscape's recommended directory structure.

For disable schema checking, select the default unless you think you may have a problem with old schema. You are strongly recommended to clean up your directory schema before you import your LDIF to the Directory Server. Further, you should NOT run a production Directory Server with schema checking turned off.

For administration port number, enter a value that is not in use. Be sure to record this value.

For the IP address to bind to, select the default unless you want the Administration Server to listen to some other IP address (this may be necessary for some multi-homed systems). If so, then enter that other IP address.

For Server Administrator ID, enter the user ID and password that you will use to authenticate to the Administration Server when the Directory Server is not running. For best results, this user ID/password pair should be the same as the user ID/password that you specified for the Configuration Directory Administrator.

For the user you want to run the Administration Server as, enter Root. This is the default. For information on why you should run the Administration Server as root, see "Decide Which User and Group to Run Netscape Servers As (Unix only)".

The server is then installed and started. You still must initialize your consumer server(s) from the supplier server's directory tree. See the Netscape Directory Server Administrator's Guide for information on consumer creation.

SIR Supplier Server Installs on Windows NT

If you have not already done so, download the product binaries file to the to the machine on which you want to install the server.

Unzip the product binaries files and run the setup program.

When you are asked what you would like to install, select the default, Netscape Servers.

When you are asked what type of installation you would like to perform, select Custom Installation.

For server root, enter a full path to the location where you want to install your server. The location that you enter must be some directory other than the directory from which you are running setup. If the directory that you specify does not exist, setup creates it for you.

For Netscape Server Family Core Components, Netscape Directory Suite, and NT Synchronization Service, select all the components except for the NT Synchronization Service. (NT Synchronization Service installation is described separately in Chapter 5, "Installing and Configuring the Synch Service."

For configuration directory, select the default. However, if this directory will not host your o=NetscapeRoot tree then select "Use existing configuration Directory Server." If you select this option, then the configuration directory must exist before you can continue this installation.

For Directory to store data, select the default ("Store data in this Directory Server.")

For server identifier, enter some unique value (normally the default is sufficient). This value is used as part of the name of the directory in which the Directory Server instance is installed. For example, if your machine's host name is phonebook then this name is the default and selecting it will cause the Directory Server instance to be installed into a directory labeled slapd-phonebook.

For a directory suffix, enter a distinguished name meaningful to your enterprise. This string is used to form the name of all your organization's directory entries. Therefore, pick some name that is representative of your organization. It is recommended that you pick a suffix that corresponds to your internet DNS name.

For example, if your organization uses the DNS name airius.com, then enter o=airius.com here.

For the Directory Server port, select the default (389) unless you already have another application using that port.

For Configuration Directory Administrator ID and password, enter the name and password that you log in as when you want to authenticate to the console with full privileges.

For Administration Domain, enter the domain that you want this server to belong to. The name that you enter should be a unique string that is descriptive of the organization responsible for administering the domain. For information on administration domains, see "Determine the Administration Domain".

For Directory Manager DN, enter the distinguished name that you will use when managing the contents of your directory with unlimited privileges (in former releases of the Directory Server, the Directory Manager was known as the root DN). This is the entry that you bind to the directory as when you want access control to be ignored. This distinguished name can be short and does not have to conform to any suffix configured for your directory. It also should not correspond to an actual entry stored in your directory. Examples of possible directory manager DNs are:

cn=Directory Manager

cn=root

uid=admin

For Directory Manager password, enter a value that is at least 8 characters long.

For Configure server for replication, select "A supplier that will push updates to another server (SIR)".

For changelog suffix and changelog database directory, select the defaults. This is the suffix used by the directory tree that will store all changes made to the supplier server, and the location of the database used to store these changes. These changes must be tracked so that the supplier can know how to update it's consumer servers.

For host and port, enter the fully qualified DNS name and port of the machine where a consumer server will be running.

For the Bind As and password, enter the supplier DN and password for the consumer server. For example, cn=supplier. By entering this value here, you are only telling the supplier server how to bind to the consumer server; you are not actually setting up the supplier DN on the consumer server. For information on how to set up a supplier DN, see the Netscape Directory Server Administrator's Guide.

For the Root of Replication, enter the distinguished name of the root point of the directory tree that this consumer server will receive. That is, if you are replicating the tree under ou=people, o=airius.com, then enter that DN here.

For replication days, specify the days of the week on which replication can begin. To replicate every day of the week, select the default (all).

For replication time, enter the range of time during which replication can begin. To allow replication to begin at any time during the day, select the default (all day).

For replication time, enter the range of time during which replication can begin. Specify times in 24 hour format. For example, 0100-2300 allows replication to begin anytime between 1 a.m. and 11 p.m. To allow replication to begin any time during the day, select the default (all day).

At this point the setup program will attempt to bind to your consumer server using the DN and password that you supplied. If the consumer server is not running, or if you have not yet created the supplier DN on the consumer server, setup will indicate that it could not bind to the consumer server. However, the setup program will continue. Once your supplier server has been installed, you should go to your consumer server and create the supplier DN required for replication purposes.