The sesman tools sesrun and sesadmin now use the separate
authentication/authorization (AA) interface introduced to
sesman by the previous comment.
sesrun can use either password or UDS authentication. With some
limitations, this can allow for automatic creation of sessions for local
users without a password being needed.
sesadmin now operates using UDS logins only and so a username and
password are not required. To use sesadmin for another user, use
su/sudo/doas to authenticate as the other user.
Update sesman to cope with separate authentication/authorization (AA) and
command processing.
Also, internally users are now tracked by UID rather thn username.
This addresses a problem found by some users using federated naming
services (e.g. Active Directory) where the same user can be referred to
in more than one way. See https://github.com/neutrinolabs/xrdp/issues/1823
The separation of AA in this way allows for multiple attempts to be made
on one connection to get a password right. This addresses MaxLoginRetry
not working (https://github.com/neutrinolabs/xrdp/issues/1739)
The previous commit introduced a new interface for the auth modules. This
commit simply updates the other auth modules to use the new interface.
The basic auth module is also updated so that if a user has a shadow
password entry indicated, but the shadow entry cannot be found, an error
is logged rather than silently succeeding.
The BSD authentication module is also updated to allow it to be
compiled on a Linux system for basic testing.
An extra method auth_uds() is added to the PAM module to
allow a 'struct auth_info' to be created for a UDS login. The PAM stack
is used to check the UDS user can be authorized.
Also, an error code is returned from the auth module rather than a
simple boolean. This allows a more complete status to be communicated
to the user. See https://github.com/neutrinolabs/xrdp/discussions/1921
and also #909 and #642
Moving to a uid_t to store the user information makes a lot
of sense. When doing this, we need a function to get information
about a user from the uid_t
As well as creating the function g_getuser_info_by_uid() we also
rename g_getuser_info() to g_getuser_info_by_name() and make the
parameter ordering more usual.
When using PAM authentication, a copy is made of the username and password in the auth_info structure.
The password copy is not cleared from memory when the structure is deallocated. This could mean the password is revealed to an attacker from a coredump.
One solution is to clear the password when the struct is deallocated. However, the username and password in the auth_info struct are only required for the duration of the PAM conversation function. A better solution is to remove the username and password from the auth_info struct entirely, and just use pointers for the duration of the time the callback function is used.
The loadable sesman authentication modules use different types for the
authentication handle returned from auth_userpass(). The PAM module
uses a pointer, and the other modules use (effectively) a boolean. Within
sesman itself, a long or tbus (intptr_t) is used.
This PR replaces all of these types with a pointer to an incomplete type.
Consequently:-
- A single better-labelled type is used it all places within sesman so
it's more obvious what's being handled.
- There is no need to cast the authentication handle within the PAM
module to a long and back again.
- The compiler can check function signatures between auth.h and the
various verify modules.
This may throw a warning with clang-15+ when devel logs are disabled
Fixes
../../../xrdp-0.9.19/sesman/chansrv/chansrv.c:198:9: error: variable 'count' set but not used [-Werror,-Wunused-but-set-variable]
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Made session allocation policies more readable and maintainable.
The 'C' policy which was confusing before has been replaced with the
'Separate' keyword. This is a public interface change, but is unlikely
to affect many users.
The logging in session_get_bydata() is substantially improved, making
it far easier to spot why sessions are getting matched or not matched.
The connected client is currently described in two places in
the xrdp_client_info structure:-
1) In the connection_description field. This was introduced as
field client_ip by commit d797b2cf49
for xrdp v0.6.0
2) In the client_addr and client_port fields introduced by commit
25369460a1 for xrdp v0.8.0
This commit unifies these two sets of fields into a single
set of fields describing the connection IP and port (for
AF_INET/AF_INET6 connections only) and a connection description
for all connection types.
The code in os_calls to provide client logging has been simplified
somewhat which should make it easier to add new connection types (e.g.
AF_VSOCK).
The old connection_description field used to be passed to sesman to
inform sesman of the IP address of the client, and also to provide
a string for 'C' field session policy matching. 'C' field session policy
matching does not actually need this string (see #2239), and so now only
the IP field is passed to sesman.
The code in xrdp_mm.c to connect to chansrv over a TCP socket has
been removed, with the move to UDS. This PR simply removes the
chansrv TCP listening code. Without doing this, some configurations
result in a failure of xrdp to connect to chansrv.
The TCP socket implementation of sesman has a number of limitations,
namely that it is affected by firewalls, and also that determining the
user on the other end requires a full authentication process.
The advantage of the TCP socket is that sesman and xrdp can be run on
separate machines. This is however not supported by the xorgxrdp
backend (shared memory), and is insecure, in that passwords are sent
in-the-clear, and the connection is susceptible to MitM attacks. This
architecture has been deprecated in release notes since xrdp v0.9.17,
and although it will continue to be supported in any further releases
in the x0.9.x series, it will not be supported in the next major
version.
When sesman used a standard TCP socket, we were guaranteed only one copy
of sesman could run on on address, as standard TCP listening rules
enforced this. This isn't the case with Unix Domain sockets. This
module implements a locking mechanism for a UDS which emulates the
standard TCP socket behaviour.
This is required for PAM systems that depend on group membership being
available during PAM processing. This is used by pam_group on FreeBSD
and pam_group on Linux-PAM, although the functionality of both is
different.
