Searching session data on a central node in a cluster
The central search functionality is available when your deployment consists of two or more instances of One Identity Safeguard for Privileged Sessions (SPS) organized into a cluster. When you have a cluster of nodes set up, you have the possibility to search all session data recorded by all nodes in the cluster on a single node. This is achieved by assigning roles to the individual nodes in your cluster: you can set up one of your Safeguard for Privileged Sessions nodes to be the Search Master and the rest of the nodes to be Search Minions. Search Minions send session data that they record to the Search Master, and the Search Master acts as a central search node.
To set up your environment for central searching, complete the following steps:
- Enable cluster management on the nodes that you want to be part of your cluster.
- Build a cluster.
-
Assign roles to nodes in your cluster.
Familiarize yourself with:
- The available search roles before assigning them to nodes. For more information, see Cluster roles.
- Managing a central search configuration. For more information, see Managing a cluster with central search configuration and configuration synchronization.
Once you have your cluster set up and the appropriate roles assigned, you can start searching session data using the Search interface.
|
|
NOTE:
Central search is not available on the Search (classic) interface. |
Limitations of the central search functionality
Currently, the central search functionality comes with the following limitations:
-
Session data recorded by a node before it was joined to the cluster will not be searchable centrally. Only session data recorded after the node has been joined to the cluster is available for central search.
-
You cannot to run the indexer process on unindexed sessions after assigning the Search Master role to a node. Make sure all important sessions are indexed before assigning the Search Master role to the node.
-
The Search Master node cannot run internal indexer processes, nor does it receive connections from external indexers. Indexers work only with Search Minion nodes.
-
User behavior analysis provided by One Identity Safeguard for Privileged Analytics is not available.
-
It is not possible to replay audit trail files in your browser from the Search Master node.
-
When near real-time indexing is configured on a Search Minion node, while session data from active connections is visible on the Search interface of the Search Master node, it is not possible to:
-
export the audit trail of an active connection,
-
follow an active connection, and
-
terminate an active connection.
Note, however, that you can terminate the active, ongoing connection on the Search Minion node that is recording the connection in question.
-
-
A reliable, high-bandwidth connection is required between the nodes. Small loss of connection is handled well but if the connection between the Search Minions and the Search Master is lost for a longer period of time, the Search Minions will stop accepting new connections until the connection is repaired. Data is automatically pushed to the Search Master after the connection is restored.
|
|
NOTE:
Search Minion nodes do not send the files storing the audit trails to the Search Master node. When a user clicks |
, the Search Master node streams the trail files to the user from the original Search Minion node that recorded the sessions. If a Search Minion node does not have a backup policy set up and an error occurs that causes data loss, then session data recorded by that node will not be available.Advanced authentication and authorization techniques
This section describes the advanced authentication and authorization techniques available in One Identity Safeguard for Privileged Sessions.
-
For details on creating usermapping policies, see Configuring usermapping policies.
-
For details on configuring gateway authentication, see Configuring gateway authentication.
-
For details on configuring four-eyes authorization and real-time monitoring, see Configuring four-eyes authorization.
-
For details on configuring Credential Stores, see Using credential stores for server-side authentication.
Configuring usermapping policies
For SSH, RDP, Telnet, and Citrix ICA connections, usermapping policies can be defined. A usermapping policy describes who can use a specific username to access the remote server: only members of the specified local or LDAP usergroups (for example, administrators) can use the specified username (for example, root) on the server.
|
|
Caution:
In SSH connections, the users must use the following as their username: gu=username@remoteusername, where username is the username used in the LDAP directory, One Identity Safeguard for Privileged Sessions will use this username to determine their group memberships, and remoteusername is the username they will use on the remote server. For example, to access the example.com server as root, use: gu=yourldapusername@root@example.com For the username of SSH users, only valid UTF-8 strings are allowed. |
|
|
Caution:
In Telnet connections, usermapping policy works only if Extract username from the traffic is enabled. For details, see Extracting username from Telnet connections. |
When configuring ICA connections, also consider the following:
|
|
Caution:
If the clients are accessing a remote application or desktop that is shared for Anonymous users (that is, the Users properties of the application is set to Allow anonymous users in the Citrix Delivery Services Console), the actual remote session will be running under an Anonymous account name (for example, Anon001, Anon002, and so on), not under the username used to access the remote server. Therefore, you need to enable usermapping to the Anon* usernames. To accomplish this, create a usermapping policy and set the Username on the server option to Anon*, and the Groups option to *, then use this usermapping policy in your ICA connections. For details on using usermapping policies, see Configuring usermapping policies. |
|
|
NOTE:
Starting from SPS version 3.2, usermapping is possible only when gateway authentication is used as well. |
To configure usermapping
-
Navigate to Policies > Usermapping Policies.
Figure 246: Policies > Usermapping Policies — Configuring usermapping policies
-
Click
to create a new policy, and enter a name for the policy.
-
Click
-
Select Groups, click
-
If you have an LDAP Server set in the connection policy where you will use usermapping, enter the name of the local or LDAP usergroup (for example admins) whose members will be permitted to use the remote username. For details on LDAP authentication, see Authenticating users to an LDAP server.
NOTE: The LDAP server configured in the connection policy is not necessarily the same as the LDAP server used to authenticate the users accessing the SPS web interface.
-
If you do not authenticate the connections to an LDAP server, enter the name of the userlist whose members will be permitted to use the remote username. For details on using userlists, see Creating and editing user lists.
Repeat this step to add further groups if needed.
-
-
Repeat steps 3-4 to add further usernames if needed.
-
To permit other users, who are not explicitly listed in the Usermapping Policy access the remote servers, select the Allow other unmapped usernames option. Note that these users must use the same username on the SPS gateway and the remote server.
-
Click
.
-
Navigate to the Connections page of the traffic (for example to SSH Control > Connections), and select the connection policy to modify.
-
Select the usermapping policy created in Step 2 from the Usermapping policy field.
-
Click
NOTE: For RDP connections, usermapping is possible only when gateway authentication is used as well. When configuring usermapping for RDP connections, proceed to Configuring out-of-band gateway authentication and configure gateway authentication.
Configuring gateway authentication
When gateway authentication is required for a connection, the user must authenticate on One Identity Safeguard for Privileged Sessions (SPS) as well. This additional authentication can be performed:
-
Out-of-band, on the SPS web interface, for every protocol.
-
Inband, using the incoming connection, for the SSH, Telnet, and RDP protocols.
For details about the concepts of gateway authentication, see The gateway authentication process. You can use gateway authentication to authenticate the real person when the user is using a shared account to access the target server.
|
|
NOTE:
For SSH, Telnet, and RDP connections, gateway authentication can be performed also inband, without having to access the SPS web interface.
|
|
|
NOTE:
Gateway authentication can be used together with other advanced authentication and authorization techniques like four-eyes authorization, client- and server-side authentication, and so on. |
|
|
Caution:
If the username used within the protocol to access the remote server is different from the username used to perform gateway authentication (for example, because the user uses a shared account in the remote server, but a personal account for gateway authentication), usermapping must be configured for the connection. For details on usermapping, see Configuring usermapping policies. |
|
|
NOTE:
To configure a credential store for gateway authentication, see Using credential stores for server-side authentication. |