Access control in SPS is based on groups. Whenever a user needs to access a protected resource, like navigating to a configuration page on the SPS web interface, or opening a channel in a connection, SPS checks the access control list associated with the resource in question.
The access control lists grant access to groups. Therefore, SPS needs to determine which groups the user is a member of to evaluate the access rules.
When you configure SPS to use an LDAP backend, SPS will:
-
Identify the user. For more information, see User identification below.
-
Determine the relevant groups the user is a member of. For more information, see Group membership resolution below.
User identification
SPS works with plain usernames, for example, administrator. This must be unambiguously resolved to an LDAP user object in order to determine the user’s groups. If a user identification returns multiple results, SPS treats this as an error, and access to the user in question is denied.
Only the user object returned in this phase is used for group membership checks, and not the original plain username.
User resolution depends on the type of the backend (POSIX or Active Directory).
For more information, see the backend-specific sections below.
Group membership resolution
SPS works with plain group names, for example, superusers. For group membership checks, SPS looks up a relevant group object in LDAP and checks if the user object returned during user identification is a member of that group. Since some of the group object’s attributes are always used for group membership checks, the group object must also exist in LDAP.
Group membership resolution depends on the LDAP backend type.
For more information, see the backend-specific sections below.