sin/samba-mirror
1
0
Fork 0
mirror of https://github.com/samba-team/samba.git synced 2025-01-28 17:47:29 +03:00
Code
samba-mirror/source4/smb_server/sesssetup.c

391 lines
12 KiB
C
Raw Normal View History

first public release of samba4 code (This used to be commit b0510b5428b3461aeb9bbe3cc95f62fc73e2b97f)
2003-08-13 01:53:07 +00:00
/*
Unix SMB/CIFS implementation.
handle SMBsessionsetup
r6270: Move the VUID handling to a IDR tree. This should avoid O(n) behaviour on session setups, and because we no longer need do deal with the linked list as much, the code is much simpiler too. We may be able to compleatly remove the tid and vuid linked lists, but I need to check. This patch also tries to clean up the VUID handling and session setups in general. To avoid security issues, we now have a distinction between VUIDs allocated for the session setup (to tie togeather the multiple round trips) and those used after authentication. Andrew Bartlett (This used to be commit 3e5775146d9ce6f0ac43aecae7e899b5324399ad)
2005-04-10 07:39:51 +00:00
Copyright (C) Andrew Tridgell 1998-2001
Copyright (C) Andrew Bartlett <abartlet@samba.org> 2001-2005
Copyright (C) Jim McDonough 2002
Copyright (C) Luke Howard 2003
first public release of samba4 code (This used to be commit b0510b5428b3461aeb9bbe3cc95f62fc73e2b97f)
2003-08-13 01:53:07 +00:00
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
*/
#include "includes.h"
r4819: its just not my day today .... (This used to be commit e54a97f8a67a04427b36cb4afac204c4e5f4502a)
2005-01-18 10:10:35 +00:00
#include "version.h"
r3453: - split out the auth and popt includes - tidied up some of the system includes - moved a few more structures back from misc.idl to netlogon.idl and samr.idl now that pidl knows about inter-IDL dependencies (This used to be commit 7b7477ac42d96faac1b0ff361525d2c63cedfc64)
2004-11-02 02:57:18 +00:00
#include "auth/auth.h"
r3466: split out request.h, signing.h, and smb_server.h (This used to be commit 7c4e6ebf05790dd6e29896dd316db0fff613aa4e)
2004-11-02 07:18:24 +00:00
#include "smb_server/smb_server.h"
r5102: This is a major simplification of the logic for controlling top level servers in smbd. The old code still contained a fairly bit of legacy from the time when smbd was only handling SMB connection. The new code gets rid of all of the smb_server specific code in smbd/, and creates a much simpler infrastructures for new server code. Major changes include: - simplified the process model code a lot. - got rid of the top level server and service structures completely. The top level context is now the event_context. This got rid of service.h and server.h completely (they were the most confusing parts of the old code) - added service_stream.[ch] for the helper functions that are specific to stream type services (services that handle streams, and use a logically separate process per connection) - got rid of the builtin idle_handler code in the service logic, as none of the servers were using it, and it can easily be handled by a server in future by adding its own timed_event to the event context. - fixed some major memory leaks in the rpc server code. - added registration of servers, rather than hard coding our list of possible servers. This allows for servers as modules in the future. - temporarily disabled the winbind code until I add the helper functions for that type of server - added error checking on service startup. If a configured server fails to startup then smbd doesn't startup. - cleaned up the command line handling in smbd, removing unused options (This used to be commit cf6a46c3cbde7b1eb1b86bd3882b953a2de3a42e)
2005-01-30 00:54:57 +00:00
#include "smbd/service_stream.h"
r6270: Move the VUID handling to a IDR tree. This should avoid O(n) behaviour on session setups, and because we no longer need do deal with the linked list as much, the code is much simpiler too. We may be able to compleatly remove the tid and vuid linked lists, but I need to check. This patch also tries to clean up the VUID handling and session setups in general. To avoid security issues, we now have a distinction between VUIDs allocated for the session setup (to tie togeather the multiple round trips) and those used after authentication. Andrew Bartlett (This used to be commit 3e5775146d9ce6f0ac43aecae7e899b5324399ad)
2005-04-10 07:39:51 +00:00
#include "libcli/nbt/libnbt.h"
first public release of samba4 code (This used to be commit b0510b5428b3461aeb9bbe3cc95f62fc73e2b97f)
2003-08-13 01:53:07 +00:00
/*
setup the OS, Lanman and domain portions of a session setup reply
*/
r1280: rename struct request_context to smbsrv_request metze (This used to be commit a85d2db5826a84b812ea5162a11f54edd25f74e3)
2004-06-28 08:39:00 +00:00
static void sesssetup_common_strings(struct smbsrv_request *req,
first public release of samba4 code (This used to be commit b0510b5428b3461aeb9bbe3cc95f62fc73e2b97f)
2003-08-13 01:53:07 +00:00
char **os, char **lanman, char **domain)
{
r2249: got rid of some more mem_ctx elements in structures (This used to be commit 21ef338cbbe96acc8594ffc550ef60c6a40fb951)
2004-09-08 05:39:06 +00:00
(*os) = talloc_asprintf(req, "Unix");
(*lanman) = talloc_asprintf(req, "Samba %s", SAMBA_VERSION_STRING);
(*domain) = talloc_asprintf(req, "%s", lp_workgroup());
first public release of samba4 code (This used to be commit b0510b5428b3461aeb9bbe3cc95f62fc73e2b97f)
2003-08-13 01:53:07 +00:00
}
/*
handler for old style session setup
*/
r1280: rename struct request_context to smbsrv_request metze (This used to be commit a85d2db5826a84b812ea5162a11f54edd25f74e3)
2004-06-28 08:39:00 +00:00
static NTSTATUS sesssetup_old(struct smbsrv_request *req, union smb_sesssetup *sess)
first public release of samba4 code (This used to be commit b0510b5428b3461aeb9bbe3cc95f62fc73e2b97f)
2003-08-13 01:53:07 +00:00
{
NTSTATUS status;
r1019: Push the auth subsystem away from using typedef, and over to the 'all goodness and light' struct ;-) Break apart the auth subsystem's return strucutres, into the parts that a netlogon call cares about, and the parts that are for a local session. This is the 'struct session_info' and it will almost completly replace the current information stored on a vuid, but be generic to all login methods (RPC over TCP, for example). Andrew Bartlett (This used to be commit d199697014d9562f9439a30b950fda798c5ef419)
2004-06-05 01:39:08 +00:00
struct auth_usersupplied_info *user_info = NULL;
struct auth_serversupplied_info *server_info = NULL;
struct auth_session_info *session_info;
r6270: Move the VUID handling to a IDR tree. This should avoid O(n) behaviour on session setups, and because we no longer need do deal with the linked list as much, the code is much simpiler too. We may be able to compleatly remove the tid and vuid linked lists, but I need to check. This patch also tries to clean up the VUID handling and session setups in general. To avoid security issues, we now have a distinction between VUIDs allocated for the session setup (to tie togeather the multiple round trips) and those used after authentication. Andrew Bartlett (This used to be commit 3e5775146d9ce6f0ac43aecae7e899b5324399ad)
2005-04-10 07:39:51 +00:00
struct smbsrv_session *smb_sess;
r8700: Propmted by tridge's need to do plaintext auth in ejs, rework the user_info strcture in auth/ This moves it to a pattern much like that found in ntvfs, with functions to migrate between PAIN, HASH and RESPONSE passwords. Instead of make_user_info*() functions, we simply fill in the control block in the callers, per recent dicussions on the lists. This removed a lot of data copies as well as error paths, as we can grab much of it with talloc. Andrew Bartlett (This used to be commit ecbd2235a3e2be937440fa1dc0aecc5a047eda88)
2005-07-22 04:10:07 +00:00
const char *remote_machine = NULL;
r4620: - add interface functions to the auth subsystem so that callers doesn't need to use function pointers anymore - make the module init much easier - a lot of cleanups don't try to read the diff in auth/ better read the new files it passes test_echo.sh and test_rpc.sh abartlet: please fix spelling fixes metze (This used to be commit 3c0d16b8236451f2cfd38fc3db8ae2906106d847)
2005-01-09 12:55:25 +00:00
TALLOC_CTX *mem_ctx;
r6270: Move the VUID handling to a IDR tree. This should avoid O(n) behaviour on session setups, and because we no longer need do deal with the linked list as much, the code is much simpiler too. We may be able to compleatly remove the tid and vuid linked lists, but I need to check. This patch also tries to clean up the VUID handling and session setups in general. To avoid security issues, we now have a distinction between VUIDs allocated for the session setup (to tie togeather the multiple round trips) and those used after authentication. Andrew Bartlett (This used to be commit 3e5775146d9ce6f0ac43aecae7e899b5324399ad)
2005-04-10 07:39:51 +00:00
sess->old.out.vuid = UID_FIELD_INVALID;
sess->old.out.action = 0;
r4620: - add interface functions to the auth subsystem so that callers doesn't need to use function pointers anymore - make the module init much easier - a lot of cleanups don't try to read the diff in auth/ better read the new files it passes test_echo.sh and test_rpc.sh abartlet: please fix spelling fixes metze (This used to be commit 3c0d16b8236451f2cfd38fc3db8ae2906106d847)
2005-01-09 12:55:25 +00:00
mem_ctx = talloc_named(req, 0, "OLD session setup");
NT_STATUS_HAVE_NO_MEMORY(mem_ctx);
r1291: rename struct smbsrv_context to smbsrv_connection because this is the connection state per transport layer (tcp) connection I also moved the substructs directly into smbsrv_connection, because they don't need a struct name and we should allway pass the complete smbsrv_connection struct into functions metze (This used to be commit 60f823f201fcedf5473008e8453a6351e73a92c7)
2004-06-29 07:40:14 +00:00
if (!req->smb_conn->negotiate.done_sesssetup) {
req->smb_conn->negotiate.max_send = sess->old.in.bufsize;
first public release of samba4 code (This used to be commit b0510b5428b3461aeb9bbe3cc95f62fc73e2b97f)
2003-08-13 01:53:07 +00:00
}
r2542: I really don't like the 'substitute' code, and I particularly don't like it in the mainline code (outside the smb.conf magic). We will need to have a more useful 'helper' routine for this, but for now we at least get a reliable IP address. Also remove the unused 'socket' structure in the smb server - it seems to have been replaced by the socket library. Andrew Bartlett (This used to be commit d8fd19a2020da6cce691c0db2b00f42e31d672cc)
2004-09-22 23:50:28 +00:00
r10946: Use the right name for the remote workstation, and always initialise it. Should fix a valgrind error volker is seeing. Andrew Bartlett (This used to be commit 11957c5f37fe0a0be465a9ce9d6d256724c5951c)
2005-10-12 22:25:51 +00:00
if (req->smb_conn->negotiate.calling_name) {
remote_machine = req->smb_conn->negotiate.calling_name->name;
r8700: Propmted by tridge's need to do plaintext auth in ejs, rework the user_info strcture in auth/ This moves it to a pattern much like that found in ntvfs, with functions to migrate between PAIN, HASH and RESPONSE passwords. Instead of make_user_info*() functions, we simply fill in the control block in the callers, per recent dicussions on the lists. This removed a lot of data copies as well as error paths, as we can grab much of it with talloc. Andrew Bartlett (This used to be commit ecbd2235a3e2be937440fa1dc0aecc5a047eda88)
2005-07-22 04:10:07 +00:00
}
if (!remote_machine) {
remote_machine = socket_get_peer_addr(req->smb_conn->connection->socket, mem_ctx);
}
user_info = talloc(req->smb_conn, struct auth_usersupplied_info);
if (!user_info) {
r3361: Allow Samba4 (I'm interested in ntlm_auth in particular) to use Samba3's winbind. This is also the start of domain membership code in Samba4, as we now (partially) parse the info3, and use it like Samba3 does. Andrew Bartlett (This used to be commit c1b7303c1c7d9fb815006c3bd2af20a0010d15a8)
2004-10-29 09:15:41 +00:00
talloc_free(mem_ctx);
r8700: Propmted by tridge's need to do plaintext auth in ejs, rework the user_info strcture in auth/ This moves it to a pattern much like that found in ntvfs, with functions to migrate between PAIN, HASH and RESPONSE passwords. Instead of make_user_info*() functions, we simply fill in the control block in the callers, per recent dicussions on the lists. This removed a lot of data copies as well as error paths, as we can grab much of it with talloc. Andrew Bartlett (This used to be commit ecbd2235a3e2be937440fa1dc0aecc5a047eda88)
2005-07-22 04:10:07 +00:00
return NT_STATUS_NO_MEMORY;
first public release of samba4 code (This used to be commit b0510b5428b3461aeb9bbe3cc95f62fc73e2b97f)
2003-08-13 01:53:07 +00:00
}
r8700: Propmted by tridge's need to do plaintext auth in ejs, rework the user_info strcture in auth/ This moves it to a pattern much like that found in ntvfs, with functions to migrate between PAIN, HASH and RESPONSE passwords. Instead of make_user_info*() functions, we simply fill in the control block in the callers, per recent dicussions on the lists. This removed a lot of data copies as well as error paths, as we can grab much of it with talloc. Andrew Bartlett (This used to be commit ecbd2235a3e2be937440fa1dc0aecc5a047eda88)
2005-07-22 04:10:07 +00:00
user_info->mapped_state = False;
user_info->flags = 0;
user_info->client.account_name = sess->old.in.user;
user_info->client.domain_name = sess->old.in.domain;
user_info->workstation_name = remote_machine;
user_info->remote_host = socket_get_peer_addr(req->smb_conn->connection->socket, mem_ctx);
user_info->password_state = AUTH_PASSWORD_RESPONSE;
user_info->password.response.lanman = sess->old.in.password;
user_info->password.response.lanman.data = talloc_steal(user_info, sess->old.in.password.data);
user_info->password.response.nt = data_blob(NULL, 0);
first public release of samba4 code (This used to be commit b0510b5428b3461aeb9bbe3cc95f62fc73e2b97f)
2003-08-13 01:53:07 +00:00
r4620: - add interface functions to the auth subsystem so that callers doesn't need to use function pointers anymore - make the module init much easier - a lot of cleanups don't try to read the diff in auth/ better read the new files it passes test_echo.sh and test_rpc.sh abartlet: please fix spelling fixes metze (This used to be commit 3c0d16b8236451f2cfd38fc3db8ae2906106d847)
2005-01-09 12:55:25 +00:00
status = auth_check_password(req->smb_conn->negotiate.auth_context,
mem_ctx, user_info, &server_info);
first public release of samba4 code (This used to be commit b0510b5428b3461aeb9bbe3cc95f62fc73e2b97f)
2003-08-13 01:53:07 +00:00
if (!NT_STATUS_IS_OK(status)) {
r3361: Allow Samba4 (I'm interested in ntlm_auth in particular) to use Samba3's winbind. This is also the start of domain membership code in Samba4, as we now (partially) parse the info3, and use it like Samba3 does. Andrew Bartlett (This used to be commit c1b7303c1c7d9fb815006c3bd2af20a0010d15a8)
2004-10-29 09:15:41 +00:00
talloc_free(mem_ctx);
r4620: - add interface functions to the auth subsystem so that callers doesn't need to use function pointers anymore - make the module init much easier - a lot of cleanups don't try to read the diff in auth/ better read the new files it passes test_echo.sh and test_rpc.sh abartlet: please fix spelling fixes metze (This used to be commit 3c0d16b8236451f2cfd38fc3db8ae2906106d847)
2005-01-09 12:55:25 +00:00
return auth_nt_status_squash(status);
r443: Update Samba4 to the auth and NTLMSSP code from Samba3. Not all the auth code is merged - only those parts that are actually being used in Samba4. There is a lot more work to do in the NTLMSSP area, and I hope to develop that work here. There is a start on this here - splitting NTLMSSP into two parts that my operate in an async fashion (before and after the actual authentication) Andrew Bartlett (This used to be commit 5876c78806e6a6c44613a1354e8d564b427d0c9f)
2004-05-02 08:45:00 +00:00
}
r3186: Use the properties of the new talloc() system to handle the auth session info and server info structures. Andrew Bartlett (This used to be commit 5bdf391b3bc10291739f5640be9a404dbbeda273)
2004-10-25 04:19:02 +00:00
/* This references server_info into session_info */
r4620: - add interface functions to the auth subsystem so that callers doesn't need to use function pointers anymore - make the module init much easier - a lot of cleanups don't try to read the diff in auth/ better read the new files it passes test_echo.sh and test_rpc.sh abartlet: please fix spelling fixes metze (This used to be commit 3c0d16b8236451f2cfd38fc3db8ae2906106d847)
2005-01-09 12:55:25 +00:00
status = auth_generate_session_info(req, server_info, &session_info);
r3361: Allow Samba4 (I'm interested in ntlm_auth in particular) to use Samba3's winbind. This is also the start of domain membership code in Samba4, as we now (partially) parse the info3, and use it like Samba3 does. Andrew Bartlett (This used to be commit c1b7303c1c7d9fb815006c3bd2af20a0010d15a8)
2004-10-29 09:15:41 +00:00
talloc_free(mem_ctx);
r1019: Push the auth subsystem away from using typedef, and over to the 'all goodness and light' struct ;-) Break apart the auth subsystem's return strucutres, into the parts that a netlogon call cares about, and the parts that are for a local session. This is the 'struct session_info' and it will almost completly replace the current information stored on a vuid, but be generic to all login methods (RPC over TCP, for example). Andrew Bartlett (This used to be commit d199697014d9562f9439a30b950fda798c5ef419)
2004-06-05 01:39:08 +00:00
if (!NT_STATUS_IS_OK(status)) {
r4620: - add interface functions to the auth subsystem so that callers doesn't need to use function pointers anymore - make the module init much easier - a lot of cleanups don't try to read the diff in auth/ better read the new files it passes test_echo.sh and test_rpc.sh abartlet: please fix spelling fixes metze (This used to be commit 3c0d16b8236451f2cfd38fc3db8ae2906106d847)
2005-01-09 12:55:25 +00:00
return auth_nt_status_squash(status);
first public release of samba4 code (This used to be commit b0510b5428b3461aeb9bbe3cc95f62fc73e2b97f)
2003-08-13 01:53:07 +00:00
}
r6270: Move the VUID handling to a IDR tree. This should avoid O(n) behaviour on session setups, and because we no longer need do deal with the linked list as much, the code is much simpiler too. We may be able to compleatly remove the tid and vuid linked lists, but I need to check. This patch also tries to clean up the VUID handling and session setups in general. To avoid security issues, we now have a distinction between VUIDs allocated for the session setup (to tie togeather the multiple round trips) and those used after authentication. Andrew Bartlett (This used to be commit 3e5775146d9ce6f0ac43aecae7e899b5324399ad)
2005-04-10 07:39:51 +00:00
smb_sess = smbsrv_register_session(req->smb_conn, session_info, NULL);
if (!smb_sess) {
r1336: check the vuid in old style sessionsetup too metze (This used to be commit 32e307857ccc99b446e7574d46b610c63ee03583)
2004-07-05 07:29:14 +00:00
return NT_STATUS_ACCESS_DENIED;
}
r6270: Move the VUID handling to a IDR tree. This should avoid O(n) behaviour on session setups, and because we no longer need do deal with the linked list as much, the code is much simpiler too. We may be able to compleatly remove the tid and vuid linked lists, but I need to check. This patch also tries to clean up the VUID handling and session setups in general. To avoid security issues, we now have a distinction between VUIDs allocated for the session setup (to tie togeather the multiple round trips) and those used after authentication. Andrew Bartlett (This used to be commit 3e5775146d9ce6f0ac43aecae7e899b5324399ad)
2005-04-10 07:39:51 +00:00
/* Ensure this is marked as a 'real' vuid, not one
* simply valid for the session setup leg */
smb_sess->finished_sesssetup = True;
/* To correctly process any AndX packet (like a tree connect)
* we need to fill in the session on the request here */
req->session = smb_sess;
sess->old.out.vuid = smb_sess->vuid;
first public release of samba4 code (This used to be commit b0510b5428b3461aeb9bbe3cc95f62fc73e2b97f)
2003-08-13 01:53:07 +00:00
sesssetup_common_strings(req,
&sess->old.out.os,
&sess->old.out.lanman,
&sess->old.out.domain);
return NT_STATUS_OK;
}
/*
handler for NT1 style session setup
*/
r1280: rename struct request_context to smbsrv_request metze (This used to be commit a85d2db5826a84b812ea5162a11f54edd25f74e3)
2004-06-28 08:39:00 +00:00
static NTSTATUS sesssetup_nt1(struct smbsrv_request *req, union smb_sesssetup *sess)
first public release of samba4 code (This used to be commit b0510b5428b3461aeb9bbe3cc95f62fc73e2b97f)
2003-08-13 01:53:07 +00:00
{
NTSTATUS status;
r8700: Propmted by tridge's need to do plaintext auth in ejs, rework the user_info strcture in auth/ This moves it to a pattern much like that found in ntvfs, with functions to migrate between PAIN, HASH and RESPONSE passwords. Instead of make_user_info*() functions, we simply fill in the control block in the callers, per recent dicussions on the lists. This removed a lot of data copies as well as error paths, as we can grab much of it with talloc. Andrew Bartlett (This used to be commit ecbd2235a3e2be937440fa1dc0aecc5a047eda88)
2005-07-22 04:10:07 +00:00
const char *remote_machine = NULL;
r8706: My previous patch oversimplied the previous change to session setup - we didn't cope with the 'anonymous NTLM under SPNEGO' login. Andrew Bartlett (This used to be commit c3cc14542e426b23e468a11803c1bab0f6fe290f)
2005-07-22 05:04:45 +00:00
struct auth_context *auth_context;
r6270: Move the VUID handling to a IDR tree. This should avoid O(n) behaviour on session setups, and because we no longer need do deal with the linked list as much, the code is much simpiler too. We may be able to compleatly remove the tid and vuid linked lists, but I need to check. This patch also tries to clean up the VUID handling and session setups in general. To avoid security issues, we now have a distinction between VUIDs allocated for the session setup (to tie togeather the multiple round trips) and those used after authentication. Andrew Bartlett (This used to be commit 3e5775146d9ce6f0ac43aecae7e899b5324399ad)
2005-04-10 07:39:51 +00:00
struct smbsrv_session *smb_sess;
r1019: Push the auth subsystem away from using typedef, and over to the 'all goodness and light' struct ;-) Break apart the auth subsystem's return strucutres, into the parts that a netlogon call cares about, and the parts that are for a local session. This is the 'struct session_info' and it will almost completly replace the current information stored on a vuid, but be generic to all login methods (RPC over TCP, for example). Andrew Bartlett (This used to be commit d199697014d9562f9439a30b950fda798c5ef419)
2004-06-05 01:39:08 +00:00
struct auth_usersupplied_info *user_info = NULL;
struct auth_serversupplied_info *server_info = NULL;
struct auth_session_info *session_info;
r4620: - add interface functions to the auth subsystem so that callers doesn't need to use function pointers anymore - make the module init much easier - a lot of cleanups don't try to read the diff in auth/ better read the new files it passes test_echo.sh and test_rpc.sh abartlet: please fix spelling fixes metze (This used to be commit 3c0d16b8236451f2cfd38fc3db8ae2906106d847)
2005-01-09 12:55:25 +00:00
TALLOC_CTX *mem_ctx;
r3361: Allow Samba4 (I'm interested in ntlm_auth in particular) to use Samba3's winbind. This is also the start of domain membership code in Samba4, as we now (partially) parse the info3, and use it like Samba3 does. Andrew Bartlett (This used to be commit c1b7303c1c7d9fb815006c3bd2af20a0010d15a8)
2004-10-29 09:15:41 +00:00
r6270: Move the VUID handling to a IDR tree. This should avoid O(n) behaviour on session setups, and because we no longer need do deal with the linked list as much, the code is much simpiler too. We may be able to compleatly remove the tid and vuid linked lists, but I need to check. This patch also tries to clean up the VUID handling and session setups in general. To avoid security issues, we now have a distinction between VUIDs allocated for the session setup (to tie togeather the multiple round trips) and those used after authentication. Andrew Bartlett (This used to be commit 3e5775146d9ce6f0ac43aecae7e899b5324399ad)
2005-04-10 07:39:51 +00:00
sess->nt1.out.vuid = UID_FIELD_INVALID;
sess->nt1.out.action = 0;
r4620: - add interface functions to the auth subsystem so that callers doesn't need to use function pointers anymore - make the module init much easier - a lot of cleanups don't try to read the diff in auth/ better read the new files it passes test_echo.sh and test_rpc.sh abartlet: please fix spelling fixes metze (This used to be commit 3c0d16b8236451f2cfd38fc3db8ae2906106d847)
2005-01-09 12:55:25 +00:00
mem_ctx = talloc_named(req, 0, "NT1 session setup");
NT_STATUS_HAVE_NO_MEMORY(mem_ctx);
first public release of samba4 code (This used to be commit b0510b5428b3461aeb9bbe3cc95f62fc73e2b97f)
2003-08-13 01:53:07 +00:00
r1291: rename struct smbsrv_context to smbsrv_connection because this is the connection state per transport layer (tcp) connection I also moved the substructs directly into smbsrv_connection, because they don't need a struct name and we should allway pass the complete smbsrv_connection struct into functions metze (This used to be commit 60f823f201fcedf5473008e8453a6351e73a92c7)
2004-06-29 07:40:14 +00:00
if (!req->smb_conn->negotiate.done_sesssetup) {
req->smb_conn->negotiate.max_send = sess->nt1.in.bufsize;
req->smb_conn->negotiate.client_caps = sess->nt1.in.capabilities;
first public release of samba4 code (This used to be commit b0510b5428b3461aeb9bbe3cc95f62fc73e2b97f)
2003-08-13 01:53:07 +00:00
}
r1731: Add server-side SPNEGO support to Samba (disabled, until SMB signing is reworked). Andrew Bartlett (This used to be commit 73ee549b8c54e93556ff0105941996e0d4de8303)
2004-08-11 18:09:40 +00:00
if (req->smb_conn->negotiate.spnego_negotiated) {
r1762: Ensure that a user (as opposed to guest) cannot login without SPNEGO, when we have negotiated SPNEGO. Andrew Bartlett (This used to be commit 07e3d2c4cd77d06c9ffaefd481ba58e4debe028c)
2004-08-12 06:58:10 +00:00
if (sess->nt1.in.user && *sess->nt1.in.user) {
r3361: Allow Samba4 (I'm interested in ntlm_auth in particular) to use Samba3's winbind. This is also the start of domain membership code in Samba4, as we now (partially) parse the info3, and use it like Samba3 does. Andrew Bartlett (This used to be commit c1b7303c1c7d9fb815006c3bd2af20a0010d15a8)
2004-10-29 09:15:41 +00:00
talloc_free(mem_ctx);
r8700: Propmted by tridge's need to do plaintext auth in ejs, rework the user_info strcture in auth/ This moves it to a pattern much like that found in ntvfs, with functions to migrate between PAIN, HASH and RESPONSE passwords. Instead of make_user_info*() functions, we simply fill in the control block in the callers, per recent dicussions on the lists. This removed a lot of data copies as well as error paths, as we can grab much of it with talloc. Andrew Bartlett (This used to be commit ecbd2235a3e2be937440fa1dc0aecc5a047eda88)
2005-07-22 04:10:07 +00:00
/* We can't accept a normal login, because we
* don't have a challenge */
return NT_STATUS_LOGON_FAILURE;
r1731: Add server-side SPNEGO support to Samba (disabled, until SMB signing is reworked). Andrew Bartlett (This used to be commit 73ee549b8c54e93556ff0105941996e0d4de8303)
2004-08-11 18:09:40 +00:00
}
r8706: My previous patch oversimplied the previous change to session setup - we didn't cope with the 'anonymous NTLM under SPNEGO' login. Andrew Bartlett (This used to be commit c3cc14542e426b23e468a11803c1bab0f6fe290f)
2005-07-22 05:04:45 +00:00
/* TODO: should we use just "anonymous" here? */
status = auth_context_create(mem_ctx, lp_auth_methods(),
&auth_context,
req->smb_conn->connection->event.ctx);
if (!NT_STATUS_IS_OK(status)) {
talloc_free(mem_ctx);
return status;
}
} else {
auth_context = req->smb_conn->negotiate.auth_context;
first public release of samba4 code (This used to be commit b0510b5428b3461aeb9bbe3cc95f62fc73e2b97f)
2003-08-13 01:53:07 +00:00
}
r10946: Use the right name for the remote workstation, and always initialise it. Should fix a valgrind error volker is seeing. Andrew Bartlett (This used to be commit 11957c5f37fe0a0be465a9ce9d6d256724c5951c)
2005-10-12 22:25:51 +00:00
if (req->smb_conn->negotiate.calling_name) {
remote_machine = req->smb_conn->negotiate.calling_name->name;
r8700: Propmted by tridge's need to do plaintext auth in ejs, rework the user_info strcture in auth/ This moves it to a pattern much like that found in ntvfs, with functions to migrate between PAIN, HASH and RESPONSE passwords. Instead of make_user_info*() functions, we simply fill in the control block in the callers, per recent dicussions on the lists. This removed a lot of data copies as well as error paths, as we can grab much of it with talloc. Andrew Bartlett (This used to be commit ecbd2235a3e2be937440fa1dc0aecc5a047eda88)
2005-07-22 04:10:07 +00:00
}
if (!remote_machine) {
remote_machine = socket_get_peer_addr(req->smb_conn->connection->socket, mem_ctx);
}
user_info = talloc(req->smb_conn, struct auth_usersupplied_info);
if (!user_info) {
talloc_free(mem_ctx);
return NT_STATUS_NO_MEMORY;
}
user_info->mapped_state = False;
user_info->flags = 0;
user_info->client.account_name = sess->nt1.in.user;
user_info->client.domain_name = sess->nt1.in.domain;
user_info->workstation_name = remote_machine;
user_info->remote_host = socket_get_peer_addr(req->smb_conn->connection->socket, mem_ctx);
user_info->password_state = AUTH_PASSWORD_RESPONSE;
user_info->password.response.lanman = sess->nt1.in.password1;
user_info->password.response.lanman.data = talloc_steal(user_info, sess->nt1.in.password1.data);
user_info->password.response.nt = sess->nt1.in.password2;
user_info->password.response.nt.data = talloc_steal(user_info, sess->nt1.in.password2.data);
r8706: My previous patch oversimplied the previous change to session setup - we didn't cope with the 'anonymous NTLM under SPNEGO' login. Andrew Bartlett (This used to be commit c3cc14542e426b23e468a11803c1bab0f6fe290f)
2005-07-22 05:04:45 +00:00
status = auth_check_password(auth_context,
mem_ctx, user_info, &server_info);
first public release of samba4 code (This used to be commit b0510b5428b3461aeb9bbe3cc95f62fc73e2b97f)
2003-08-13 01:53:07 +00:00
if (!NT_STATUS_IS_OK(status)) {
r3361: Allow Samba4 (I'm interested in ntlm_auth in particular) to use Samba3's winbind. This is also the start of domain membership code in Samba4, as we now (partially) parse the info3, and use it like Samba3 does. Andrew Bartlett (This used to be commit c1b7303c1c7d9fb815006c3bd2af20a0010d15a8)
2004-10-29 09:15:41 +00:00
talloc_free(mem_ctx);
r4620: - add interface functions to the auth subsystem so that callers doesn't need to use function pointers anymore - make the module init much easier - a lot of cleanups don't try to read the diff in auth/ better read the new files it passes test_echo.sh and test_rpc.sh abartlet: please fix spelling fixes metze (This used to be commit 3c0d16b8236451f2cfd38fc3db8ae2906106d847)
2005-01-09 12:55:25 +00:00
return auth_nt_status_squash(status);
r443: Update Samba4 to the auth and NTLMSSP code from Samba3. Not all the auth code is merged - only those parts that are actually being used in Samba4. There is a lot more work to do in the NTLMSSP area, and I hope to develop that work here. There is a start on this here - splitting NTLMSSP into two parts that my operate in an async fashion (before and after the actual authentication) Andrew Bartlett (This used to be commit 5876c78806e6a6c44613a1354e8d564b427d0c9f)
2004-05-02 08:45:00 +00:00
}
r3186: Use the properties of the new talloc() system to handle the auth session info and server info structures. Andrew Bartlett (This used to be commit 5bdf391b3bc10291739f5640be9a404dbbeda273)
2004-10-25 04:19:02 +00:00
/* This references server_info into session_info */
r4620: - add interface functions to the auth subsystem so that callers doesn't need to use function pointers anymore - make the module init much easier - a lot of cleanups don't try to read the diff in auth/ better read the new files it passes test_echo.sh and test_rpc.sh abartlet: please fix spelling fixes metze (This used to be commit 3c0d16b8236451f2cfd38fc3db8ae2906106d847)
2005-01-09 12:55:25 +00:00
status = auth_generate_session_info(mem_ctx, server_info, &session_info);
r1019: Push the auth subsystem away from using typedef, and over to the 'all goodness and light' struct ;-) Break apart the auth subsystem's return strucutres, into the parts that a netlogon call cares about, and the parts that are for a local session. This is the 'struct session_info' and it will almost completly replace the current information stored on a vuid, but be generic to all login methods (RPC over TCP, for example). Andrew Bartlett (This used to be commit d199697014d9562f9439a30b950fda798c5ef419)
2004-06-05 01:39:08 +00:00
if (!NT_STATUS_IS_OK(status)) {
r3361: Allow Samba4 (I'm interested in ntlm_auth in particular) to use Samba3's winbind. This is also the start of domain membership code in Samba4, as we now (partially) parse the info3, and use it like Samba3 does. Andrew Bartlett (This used to be commit c1b7303c1c7d9fb815006c3bd2af20a0010d15a8)
2004-10-29 09:15:41 +00:00
talloc_free(mem_ctx);
r4620: - add interface functions to the auth subsystem so that callers doesn't need to use function pointers anymore - make the module init much easier - a lot of cleanups don't try to read the diff in auth/ better read the new files it passes test_echo.sh and test_rpc.sh abartlet: please fix spelling fixes metze (This used to be commit 3c0d16b8236451f2cfd38fc3db8ae2906106d847)
2005-01-09 12:55:25 +00:00
return auth_nt_status_squash(status);
first public release of samba4 code (This used to be commit b0510b5428b3461aeb9bbe3cc95f62fc73e2b97f)
2003-08-13 01:53:07 +00:00
}
r2797: don't free the server_info before using it for anonymous connections (This used to be commit 5f5b04196c7930c91e6c00e0276f25f88181b317)
2004-10-03 07:32:08 +00:00
r6270: Move the VUID handling to a IDR tree. This should avoid O(n) behaviour on session setups, and because we no longer need do deal with the linked list as much, the code is much simpiler too. We may be able to compleatly remove the tid and vuid linked lists, but I need to check. This patch also tries to clean up the VUID handling and session setups in general. To avoid security issues, we now have a distinction between VUIDs allocated for the session setup (to tie togeather the multiple round trips) and those used after authentication. Andrew Bartlett (This used to be commit 3e5775146d9ce6f0ac43aecae7e899b5324399ad)
2005-04-10 07:39:51 +00:00
smb_sess = smbsrv_register_session(req->smb_conn, session_info, NULL);
r3361: Allow Samba4 (I'm interested in ntlm_auth in particular) to use Samba3's winbind. This is also the start of domain membership code in Samba4, as we now (partially) parse the info3, and use it like Samba3 does. Andrew Bartlett (This used to be commit c1b7303c1c7d9fb815006c3bd2af20a0010d15a8)
2004-10-29 09:15:41 +00:00
talloc_free(mem_ctx);
r6270: Move the VUID handling to a IDR tree. This should avoid O(n) behaviour on session setups, and because we no longer need do deal with the linked list as much, the code is much simpiler too. We may be able to compleatly remove the tid and vuid linked lists, but I need to check. This patch also tries to clean up the VUID handling and session setups in general. To avoid security issues, we now have a distinction between VUIDs allocated for the session setup (to tie togeather the multiple round trips) and those used after authentication. Andrew Bartlett (This used to be commit 3e5775146d9ce6f0ac43aecae7e899b5324399ad)
2005-04-10 07:39:51 +00:00
if (!smb_sess) {
r1019: Push the auth subsystem away from using typedef, and over to the 'all goodness and light' struct ;-) Break apart the auth subsystem's return strucutres, into the parts that a netlogon call cares about, and the parts that are for a local session. This is the 'struct session_info' and it will almost completly replace the current information stored on a vuid, but be generic to all login methods (RPC over TCP, for example). Andrew Bartlett (This used to be commit d199697014d9562f9439a30b950fda798c5ef419)
2004-06-05 01:39:08 +00:00
return NT_STATUS_ACCESS_DENIED;
}
r6270: Move the VUID handling to a IDR tree. This should avoid O(n) behaviour on session setups, and because we no longer need do deal with the linked list as much, the code is much simpiler too. We may be able to compleatly remove the tid and vuid linked lists, but I need to check. This patch also tries to clean up the VUID handling and session setups in general. To avoid security issues, we now have a distinction between VUIDs allocated for the session setup (to tie togeather the multiple round trips) and those used after authentication. Andrew Bartlett (This used to be commit 3e5775146d9ce6f0ac43aecae7e899b5324399ad)
2005-04-10 07:39:51 +00:00
/* Ensure this is marked as a 'real' vuid, not one
* simply valid for the session setup leg */
smb_sess->finished_sesssetup = True;
/* To correctly process any AndX packet (like a tree connect)
* we need to fill in the session on the request here */
req->session = smb_sess;
sess->nt1.out.vuid = smb_sess->vuid;
first public release of samba4 code (This used to be commit b0510b5428b3461aeb9bbe3cc95f62fc73e2b97f)
2003-08-13 01:53:07 +00:00
sesssetup_common_strings(req,
&sess->nt1.out.os,
&sess->nt1.out.lanman,
&sess->nt1.out.domain);
r1019: Push the auth subsystem away from using typedef, and over to the 'all goodness and light' struct ;-) Break apart the auth subsystem's return strucutres, into the parts that a netlogon call cares about, and the parts that are for a local session. This is the 'struct session_info' and it will almost completly replace the current information stored on a vuid, but be generic to all login methods (RPC over TCP, for example). Andrew Bartlett (This used to be commit d199697014d9562f9439a30b950fda798c5ef419)
2004-06-05 01:39:08 +00:00
r1507: fixed the handling of SMB chaining with the new server structure. You must think carefully about packet chaining when dealing with any authentication or SMB parsing issues. The particular problem here was that a chained tconX didn't get the req->session setup after an initial sesstion setup call, so the tconx used a bogus VUID. (This used to be commit 6f2a335cd623211071b01d982d4e7c69b49a5602)
2004-07-15 02:11:03 +00:00
req->session = smbsrv_session_find(req->smb_conn, sess->nt1.out.vuid);
r4620: - add interface functions to the auth subsystem so that callers doesn't need to use function pointers anymore - make the module init much easier - a lot of cleanups don't try to read the diff in auth/ better read the new files it passes test_echo.sh and test_rpc.sh abartlet: please fix spelling fixes metze (This used to be commit 3c0d16b8236451f2cfd38fc3db8ae2906106d847)
2005-01-09 12:55:25 +00:00
if (!session_info->server_info->authenticated) {
r1796: Enable server-side SPNEGO, now that I have fixed the server-side SMB signing code to be able to cope. Andrew Bartlett (This used to be commit cb74d52b563730a50e33c92d868c45ee96a598e8)
2004-08-13 00:16:57 +00:00
return NT_STATUS_OK;
r1687: Fix bogus requirement for SMB signing on guest connections. Andrew Bartlett (This used to be commit 3520af0f3d8826ac52a7fb6a658ed0924e51bbf7)
2004-08-10 04:56:44 +00:00
}
r4620: - add interface functions to the auth subsystem so that callers doesn't need to use function pointers anymore - make the module init much easier - a lot of cleanups don't try to read the diff in auth/ better read the new files it passes test_echo.sh and test_rpc.sh abartlet: please fix spelling fixes metze (This used to be commit 3c0d16b8236451f2cfd38fc3db8ae2906106d847)
2005-01-09 12:55:25 +00:00
r6270: Move the VUID handling to a IDR tree. This should avoid O(n) behaviour on session setups, and because we no longer need do deal with the linked list as much, the code is much simpiler too. We may be able to compleatly remove the tid and vuid linked lists, but I need to check. This patch also tries to clean up the VUID handling and session setups in general. To avoid security issues, we now have a distinction between VUIDs allocated for the session setup (to tie togeather the multiple round trips) and those used after authentication. Andrew Bartlett (This used to be commit 3e5775146d9ce6f0ac43aecae7e899b5324399ad)
2005-04-10 07:39:51 +00:00
if (!srv_setup_signing(req->smb_conn, &session_info->session_key, &sess->nt1.in.password2)) {
r1796: Enable server-side SPNEGO, now that I have fixed the server-side SMB signing code to be able to cope. Andrew Bartlett (This used to be commit cb74d52b563730a50e33c92d868c45ee96a598e8)
2004-08-13 00:16:57 +00:00
/* Already signing, or disabled */
return NT_STATUS_OK;
}
r4620: - add interface functions to the auth subsystem so that callers doesn't need to use function pointers anymore - make the module init much easier - a lot of cleanups don't try to read the diff in auth/ better read the new files it passes test_echo.sh and test_rpc.sh abartlet: please fix spelling fixes metze (This used to be commit 3c0d16b8236451f2cfd38fc3db8ae2906106d847)
2005-01-09 12:55:25 +00:00
r1796: Enable server-side SPNEGO, now that I have fixed the server-side SMB signing code to be able to cope. Andrew Bartlett (This used to be commit cb74d52b563730a50e33c92d868c45ee96a598e8)
2004-08-13 00:16:57 +00:00
/* Force check of the request packet, now we know the session key */
req_signing_check_incoming(req);
r6270: Move the VUID handling to a IDR tree. This should avoid O(n) behaviour on session setups, and because we no longer need do deal with the linked list as much, the code is much simpiler too. We may be able to compleatly remove the tid and vuid linked lists, but I need to check. This patch also tries to clean up the VUID handling and session setups in general. To avoid security issues, we now have a distinction between VUIDs allocated for the session setup (to tie togeather the multiple round trips) and those used after authentication. Andrew Bartlett (This used to be commit 3e5775146d9ce6f0ac43aecae7e899b5324399ad)
2005-04-10 07:39:51 +00:00
/* Unfortunetly win2k3 as a client doesn't sign the request
* packet here, so we have to force signing to start again */
r1796: Enable server-side SPNEGO, now that I have fixed the server-side SMB signing code to be able to cope. Andrew Bartlett (This used to be commit cb74d52b563730a50e33c92d868c45ee96a598e8)
2004-08-13 00:16:57 +00:00
srv_signing_restart(req->smb_conn, &session_info->session_key, &sess->nt1.in.password2);
r1687: Fix bogus requirement for SMB signing on guest connections. Andrew Bartlett (This used to be commit 3520af0f3d8826ac52a7fb6a658ed0924e51bbf7)
2004-08-10 04:56:44 +00:00
first public release of samba4 code (This used to be commit b0510b5428b3461aeb9bbe3cc95f62fc73e2b97f)
2003-08-13 01:53:07 +00:00
return NT_STATUS_OK;
}
/*
handler for SPNEGO style session setup
*/
r1280: rename struct request_context to smbsrv_request metze (This used to be commit a85d2db5826a84b812ea5162a11f54edd25f74e3)
2004-06-28 08:39:00 +00:00
static NTSTATUS sesssetup_spnego(struct smbsrv_request *req, union smb_sesssetup *sess)
first public release of samba4 code (This used to be commit b0510b5428b3461aeb9bbe3cc95f62fc73e2b97f)
2003-08-13 01:53:07 +00:00
{
r1731: Add server-side SPNEGO support to Samba (disabled, until SMB signing is reworked). Andrew Bartlett (This used to be commit 73ee549b8c54e93556ff0105941996e0d4de8303)
2004-08-11 18:09:40 +00:00
NTSTATUS status = NT_STATUS_ACCESS_DENIED;
struct smbsrv_session *smb_sess;
r8627: fixed a big memory leak in the spnego gensec code in session setup. Andrew, please check over this. What happens is this: - run the BASE-SECLEAK test - with each failed session setup using spnego a gensec ctx is leaked into the smb_conn structure - after the client disconnects these are finally cleaned up as they are all children of the connection structure - the cleanup of the millions of memory objects takes long enough that the next operation in test_posix.sh sometimes fails with a timeout Andrew, can you also look at the talloc_reference() on line 332 of sesssetup.c ? I suspect it isn't needed (I don't think it does any actual harm though) (This used to be commit b40fb6a4569ccc1fa1750a1e534e18a020764b4c)
2005-07-20 04:19:45 +00:00
struct gensec_security *gensec_ctx;
r1731: Add server-side SPNEGO support to Samba (disabled, until SMB signing is reworked). Andrew Bartlett (This used to be commit 73ee549b8c54e93556ff0105941996e0d4de8303)
2004-08-11 18:09:40 +00:00
struct auth_session_info *session_info = NULL;
uint16_t vuid;
r6270: Move the VUID handling to a IDR tree. This should avoid O(n) behaviour on session setups, and because we no longer need do deal with the linked list as much, the code is much simpiler too. We may be able to compleatly remove the tid and vuid linked lists, but I need to check. This patch also tries to clean up the VUID handling and session setups in general. To avoid security issues, we now have a distinction between VUIDs allocated for the session setup (to tie togeather the multiple round trips) and those used after authentication. Andrew Bartlett (This used to be commit 3e5775146d9ce6f0ac43aecae7e899b5324399ad)
2005-04-10 07:39:51 +00:00
sess->spnego.out.vuid = UID_FIELD_INVALID;
sess->spnego.out.action = 0;
sesssetup_common_strings(req,
&sess->spnego.out.os,
&sess->spnego.out.lanman,
&sess->spnego.out.workgroup);
r1731: Add server-side SPNEGO support to Samba (disabled, until SMB signing is reworked). Andrew Bartlett (This used to be commit 73ee549b8c54e93556ff0105941996e0d4de8303)
2004-08-11 18:09:40 +00:00
if (!req->smb_conn->negotiate.done_sesssetup) {
req->smb_conn->negotiate.max_send = sess->nt1.in.bufsize;
req->smb_conn->negotiate.client_caps = sess->nt1.in.capabilities;
}
vuid = SVAL(req->in.hdr,HDR_UID);
r6270: Move the VUID handling to a IDR tree. This should avoid O(n) behaviour on session setups, and because we no longer need do deal with the linked list as much, the code is much simpiler too. We may be able to compleatly remove the tid and vuid linked lists, but I need to check. This patch also tries to clean up the VUID handling and session setups in general. To avoid security issues, we now have a distinction between VUIDs allocated for the session setup (to tie togeather the multiple round trips) and those used after authentication. Andrew Bartlett (This used to be commit 3e5775146d9ce6f0ac43aecae7e899b5324399ad)
2005-04-10 07:39:51 +00:00
smb_sess = smbsrv_session_find_sesssetup(req->smb_conn, vuid);
if (smb_sess) {
gensec_ctx = smb_sess->gensec_ctx;
status = gensec_update(gensec_ctx, req, sess->spnego.in.secblob, &sess->spnego.out.secblob);
r1731: Add server-side SPNEGO support to Samba (disabled, until SMB signing is reworked). Andrew Bartlett (This used to be commit 73ee549b8c54e93556ff0105941996e0d4de8303)
2004-08-11 18:09:40 +00:00
} else {
r8627: fixed a big memory leak in the spnego gensec code in session setup. Andrew, please check over this. What happens is this: - run the BASE-SECLEAK test - with each failed session setup using spnego a gensec ctx is leaked into the smb_conn structure - after the client disconnects these are finally cleaned up as they are all children of the connection structure - the cleanup of the millions of memory objects takes long enough that the next operation in test_posix.sh sometimes fails with a timeout Andrew, can you also look at the talloc_reference() on line 332 of sesssetup.c ? I suspect it isn't needed (I don't think it does any actual harm though) (This used to be commit b40fb6a4569ccc1fa1750a1e534e18a020764b4c)
2005-07-20 04:19:45 +00:00
status = gensec_server_start(req, &gensec_ctx,
r7633: this patch started as an attempt to make the dcerpc code use a given event_context for the socket_connect() call, so that when things that use dcerpc are running alongside anything else it doesn't block the whole process during a connect. Then of course I needed to change any code that created a dcerpc connection (such as the auth code) to also take an event context, and anything that called that and so on .... thus the size of the patch. There were 3 places where I punted: - abartlet wanted me to add a gensec_set_event_context() call instead of adding it to the gensec init calls. Andrew, my apologies for not doing this. I didn't do it as adding a new parameter allowed me to catch all the callers with the compiler. Now that its done, we could go back and use gensec_set_event_context() - the ejs code calls auth initialisation, which means it should pass in the event context from the web server. I punted on that. Needs fixing. - I used a NULL event context in dcom_get_pipe(). This is equivalent to what we did already, but should be fixed to use a callers event context. Jelmer, can you think of a clean way to do that? I also cleaned up a couple of things: - libnet_context_destroy() makes no sense. I removed it. - removed some unused vars in various places (This used to be commit 3a3025485bdb8f600ab528c0b4b4eef0c65e3fc9)
2005-06-16 11:36:09 +00:00
req->smb_conn->connection->event.ctx);
r1731: Add server-side SPNEGO support to Samba (disabled, until SMB signing is reworked). Andrew Bartlett (This used to be commit 73ee549b8c54e93556ff0105941996e0d4de8303)
2004-08-11 18:09:40 +00:00
if (!NT_STATUS_IS_OK(status)) {
DEBUG(1, ("Failed to start GENSEC server code: %s\n", nt_errstr(status)));
return status;
}
r11200: Reposition the creation of the kerberos keytab for GSSAPI and Krb5 authentication. This pulls the creating of the keytab back to the credentials code, and removes the special case of 'use keberos keytab = yes' for now. This allows (and requires) the callers to specify the credentials for the server credentails to GENSEC. This allows kpasswdd (soon to be added) to use a different set of kerberos credentials. The 'use kerberos keytab' code will be moved into the credentials layer, as the layers below now expect a keytab. We also now allow for the old secret to be stored into the credentials, allowing service password changes. Andrew Bartlett (This used to be commit 205f77c579ac8680c85f713a76de5767189c627b)
2005-10-20 03:47:55 +00:00
gensec_set_credentials(gensec_ctx, req->smb_conn->negotiate.server_credentials);
r4355: More work from the elves on Christmas eve: - Update Samba4's kerberos code to match the 'salting' changes in Samba3 (and many other cleanups by jra). - Move GENSEC into the modern era of talloc destructors. This avoids many of the memory leaks in this code, as we now can't somehow 'forget' to call the end routine. - This required fixing some of the talloc hierarchies. - The new krb5 seems more sensitive to getting the service name right, so start actually setting the service name on the krb5 context. Andrew Bartlett (This used to be commit 278bf1a61a6da6ef955a12c13d7b1a0357cebf1f)
2004-12-24 09:54:23 +00:00
gensec_set_target_service(gensec_ctx, "cifs");
r4079: implement the gensec_have_feature() correctly by asking the backend what is actually in use metze (This used to be commit 6f3eb7bc03609108b9e0ea5676fca3d04140e737)
2004-12-06 15:44:17 +00:00
gensec_want_feature(gensec_ctx, GENSEC_FEATURE_SESSION_KEY);
r2041: Fix NTLMSSP RPC sealing, client -> win2k3 server. The bug (found by tridge) is that Win2k3 is being tighter about the NTLMSSP flags. If we don't negotiate sealing, we can't use it. We now have a way to indicate to the GENSEC implementation mechanisms what things we want for a connection. Andrew Bartlett (This used to be commit 86f61568ea44c5719f9b583beeeefb12e0c26f4c)
2004-08-25 02:25:20 +00:00
r4070: move some defines from asn_1.h to the places they belong to metze (This used to be commit ab2c2f27e1c61516e885f02bf26350f97209057a)
2004-12-05 16:29:27 +00:00
status = gensec_start_mech_by_oid(gensec_ctx, GENSEC_OID_SPNEGO);
r1731: Add server-side SPNEGO support to Samba (disabled, until SMB signing is reworked). Andrew Bartlett (This used to be commit 73ee549b8c54e93556ff0105941996e0d4de8303)
2004-08-11 18:09:40 +00:00
if (!NT_STATUS_IS_OK(status)) {
DEBUG(1, ("Failed to start GENSEC SPNEGO server code: %s\n", nt_errstr(status)));
return status;
}
r2249: got rid of some more mem_ctx elements in structures (This used to be commit 21ef338cbbe96acc8594ffc550ef60c6a40fb951)
2004-09-08 05:39:06 +00:00
status = gensec_update(gensec_ctx, req, sess->spnego.in.secblob, &sess->spnego.out.secblob);
r1747: don't segfault when the spnego mech only use one call from the client to finish metze (This used to be commit ddac5e46d42d3b2daae10107b1bcb3b138de7474)
2004-08-12 03:23:19 +00:00
}
r1731: Add server-side SPNEGO support to Samba (disabled, until SMB signing is reworked). Andrew Bartlett (This used to be commit 73ee549b8c54e93556ff0105941996e0d4de8303)
2004-08-11 18:09:40 +00:00
if (NT_STATUS_IS_OK(status)) {
DATA_BLOB session_key;
r1747: don't segfault when the spnego mech only use one call from the client to finish metze (This used to be commit ddac5e46d42d3b2daae10107b1bcb3b138de7474)
2004-08-12 03:23:19 +00:00
r6270: Move the VUID handling to a IDR tree. This should avoid O(n) behaviour on session setups, and because we no longer need do deal with the linked list as much, the code is much simpiler too. We may be able to compleatly remove the tid and vuid linked lists, but I need to check. This patch also tries to clean up the VUID handling and session setups in general. To avoid security issues, we now have a distinction between VUIDs allocated for the session setup (to tie togeather the multiple round trips) and those used after authentication. Andrew Bartlett (This used to be commit 3e5775146d9ce6f0ac43aecae7e899b5324399ad)
2005-04-10 07:39:51 +00:00
status = gensec_session_info(gensec_ctx, &session_info);
r1747: don't segfault when the spnego mech only use one call from the client to finish metze (This used to be commit ddac5e46d42d3b2daae10107b1bcb3b138de7474)
2004-08-12 03:23:19 +00:00
if (!NT_STATUS_IS_OK(status)) {
r6270: Move the VUID handling to a IDR tree. This should avoid O(n) behaviour on session setups, and because we no longer need do deal with the linked list as much, the code is much simpiler too. We may be able to compleatly remove the tid and vuid linked lists, but I need to check. This patch also tries to clean up the VUID handling and session setups in general. To avoid security issues, we now have a distinction between VUIDs allocated for the session setup (to tie togeather the multiple round trips) and those used after authentication. Andrew Bartlett (This used to be commit 3e5775146d9ce6f0ac43aecae7e899b5324399ad)
2005-04-10 07:39:51 +00:00
talloc_free(smb_sess);
r1747: don't segfault when the spnego mech only use one call from the client to finish metze (This used to be commit ddac5e46d42d3b2daae10107b1bcb3b138de7474)
2004-08-12 03:23:19 +00:00
return status;
}
r6270: Move the VUID handling to a IDR tree. This should avoid O(n) behaviour on session setups, and because we no longer need do deal with the linked list as much, the code is much simpiler too. We may be able to compleatly remove the tid and vuid linked lists, but I need to check. This patch also tries to clean up the VUID handling and session setups in general. To avoid security issues, we now have a distinction between VUIDs allocated for the session setup (to tie togeather the multiple round trips) and those used after authentication. Andrew Bartlett (This used to be commit 3e5775146d9ce6f0ac43aecae7e899b5324399ad)
2005-04-10 07:39:51 +00:00
status = gensec_session_key(gensec_ctx,
r1796: Enable server-side SPNEGO, now that I have fixed the server-side SMB signing code to be able to cope. Andrew Bartlett (This used to be commit cb74d52b563730a50e33c92d868c45ee96a598e8)
2004-08-13 00:16:57 +00:00
&session_key);
r6270: Move the VUID handling to a IDR tree. This should avoid O(n) behaviour on session setups, and because we no longer need do deal with the linked list as much, the code is much simpiler too. We may be able to compleatly remove the tid and vuid linked lists, but I need to check. This patch also tries to clean up the VUID handling and session setups in general. To avoid security issues, we now have a distinction between VUIDs allocated for the session setup (to tie togeather the multiple round trips) and those used after authentication. Andrew Bartlett (This used to be commit 3e5775146d9ce6f0ac43aecae7e899b5324399ad)
2005-04-10 07:39:51 +00:00
r1796: Enable server-side SPNEGO, now that I have fixed the server-side SMB signing code to be able to cope. Andrew Bartlett (This used to be commit cb74d52b563730a50e33c92d868c45ee96a598e8)
2004-08-13 00:16:57 +00:00
if (NT_STATUS_IS_OK(status)
r6270: Move the VUID handling to a IDR tree. This should avoid O(n) behaviour on session setups, and because we no longer need do deal with the linked list as much, the code is much simpiler too. We may be able to compleatly remove the tid and vuid linked lists, but I need to check. This patch also tries to clean up the VUID handling and session setups in general. To avoid security issues, we now have a distinction between VUIDs allocated for the session setup (to tie togeather the multiple round trips) and those used after authentication. Andrew Bartlett (This used to be commit 3e5775146d9ce6f0ac43aecae7e899b5324399ad)
2005-04-10 07:39:51 +00:00
&& session_info->server_info->authenticated
r1796: Enable server-side SPNEGO, now that I have fixed the server-side SMB signing code to be able to cope. Andrew Bartlett (This used to be commit cb74d52b563730a50e33c92d868c45ee96a598e8)
2004-08-13 00:16:57 +00:00
&& srv_setup_signing(req->smb_conn, &session_key, NULL)) {
/* Force check of the request packet, now we know the session key */
req_signing_check_incoming(req);
srv_signing_restart(req->smb_conn, &session_key, NULL);
r1731: Add server-side SPNEGO support to Samba (disabled, until SMB signing is reworked). Andrew Bartlett (This used to be commit 73ee549b8c54e93556ff0105941996e0d4de8303)
2004-08-11 18:09:40 +00:00
}
r6270: Move the VUID handling to a IDR tree. This should avoid O(n) behaviour on session setups, and because we no longer need do deal with the linked list as much, the code is much simpiler too. We may be able to compleatly remove the tid and vuid linked lists, but I need to check. This patch also tries to clean up the VUID handling and session setups in general. To avoid security issues, we now have a distinction between VUIDs allocated for the session setup (to tie togeather the multiple round trips) and those used after authentication. Andrew Bartlett (This used to be commit 3e5775146d9ce6f0ac43aecae7e899b5324399ad)
2005-04-10 07:39:51 +00:00
} else if (NT_STATUS_EQUAL(status, NT_STATUS_MORE_PROCESSING_REQUIRED)) {
r1796: Enable server-side SPNEGO, now that I have fixed the server-side SMB signing code to be able to cope. Andrew Bartlett (This used to be commit cb74d52b563730a50e33c92d868c45ee96a598e8)
2004-08-13 00:16:57 +00:00
} else {
r4620: - add interface functions to the auth subsystem so that callers doesn't need to use function pointers anymore - make the module init much easier - a lot of cleanups don't try to read the diff in auth/ better read the new files it passes test_echo.sh and test_rpc.sh abartlet: please fix spelling fixes metze (This used to be commit 3c0d16b8236451f2cfd38fc3db8ae2906106d847)
2005-01-09 12:55:25 +00:00
status = auth_nt_status_squash(status);
r6270: Move the VUID handling to a IDR tree. This should avoid O(n) behaviour on session setups, and because we no longer need do deal with the linked list as much, the code is much simpiler too. We may be able to compleatly remove the tid and vuid linked lists, but I need to check. This patch also tries to clean up the VUID handling and session setups in general. To avoid security issues, we now have a distinction between VUIDs allocated for the session setup (to tie togeather the multiple round trips) and those used after authentication. Andrew Bartlett (This used to be commit 3e5775146d9ce6f0ac43aecae7e899b5324399ad)
2005-04-10 07:39:51 +00:00
/* This invalidates the VUID of the failed login */
talloc_free(smb_sess);
return status;
}
if (!smb_sess) {
smb_sess = smbsrv_register_session(req->smb_conn,
session_info, gensec_ctx);
if (!smb_sess) {
return NT_STATUS_ACCESS_DENIED;
r2590: fixed one of the server security memory leaks. There are more :( (This used to be commit 9e1eb58e4b332e4a300e8b546a5d39bd2f7cd7a6)
2004-09-24 07:08:02 +00:00
}
r6270: Move the VUID handling to a IDR tree. This should avoid O(n) behaviour on session setups, and because we no longer need do deal with the linked list as much, the code is much simpiler too. We may be able to compleatly remove the tid and vuid linked lists, but I need to check. This patch also tries to clean up the VUID handling and session setups in general. To avoid security issues, we now have a distinction between VUIDs allocated for the session setup (to tie togeather the multiple round trips) and those used after authentication. Andrew Bartlett (This used to be commit 3e5775146d9ce6f0ac43aecae7e899b5324399ad)
2005-04-10 07:39:51 +00:00
req->session = smb_sess;
r8627: fixed a big memory leak in the spnego gensec code in session setup. Andrew, please check over this. What happens is this: - run the BASE-SECLEAK test - with each failed session setup using spnego a gensec ctx is leaked into the smb_conn structure - after the client disconnects these are finally cleaned up as they are all children of the connection structure - the cleanup of the millions of memory objects takes long enough that the next operation in test_posix.sh sometimes fails with a timeout Andrew, can you also look at the talloc_reference() on line 332 of sesssetup.c ? I suspect it isn't needed (I don't think it does any actual harm though) (This used to be commit b40fb6a4569ccc1fa1750a1e534e18a020764b4c)
2005-07-20 04:19:45 +00:00
talloc_steal(smb_sess, gensec_ctx);
r6270: Move the VUID handling to a IDR tree. This should avoid O(n) behaviour on session setups, and because we no longer need do deal with the linked list as much, the code is much simpiler too. We may be able to compleatly remove the tid and vuid linked lists, but I need to check. This patch also tries to clean up the VUID handling and session setups in general. To avoid security issues, we now have a distinction between VUIDs allocated for the session setup (to tie togeather the multiple round trips) and those used after authentication. Andrew Bartlett (This used to be commit 3e5775146d9ce6f0ac43aecae7e899b5324399ad)
2005-04-10 07:39:51 +00:00
} else {
smb_sess->session_info = talloc_reference(smb_sess, session_info);
r1731: Add server-side SPNEGO support to Samba (disabled, until SMB signing is reworked). Andrew Bartlett (This used to be commit 73ee549b8c54e93556ff0105941996e0d4de8303)
2004-08-11 18:09:40 +00:00
}
r6270: Move the VUID handling to a IDR tree. This should avoid O(n) behaviour on session setups, and because we no longer need do deal with the linked list as much, the code is much simpiler too. We may be able to compleatly remove the tid and vuid linked lists, but I need to check. This patch also tries to clean up the VUID handling and session setups in general. To avoid security issues, we now have a distinction between VUIDs allocated for the session setup (to tie togeather the multiple round trips) and those used after authentication. Andrew Bartlett (This used to be commit 3e5775146d9ce6f0ac43aecae7e899b5324399ad)
2005-04-10 07:39:51 +00:00
if (NT_STATUS_IS_OK(status)) {
/* Ensure this is marked as a 'real' vuid, not one
* simply valid for the session setup leg */
smb_sess->finished_sesssetup = True;
}
sess->spnego.out.vuid = smb_sess->vuid;
r1731: Add server-side SPNEGO support to Samba (disabled, until SMB signing is reworked). Andrew Bartlett (This used to be commit 73ee549b8c54e93556ff0105941996e0d4de8303)
2004-08-11 18:09:40 +00:00
return status;
first public release of samba4 code (This used to be commit b0510b5428b3461aeb9bbe3cc95f62fc73e2b97f)
2003-08-13 01:53:07 +00:00
}
/*
r1268: varient -> variant (This used to be commit de5984c95602ca67e8ac3139c3aa4330b74266e0)
2004-06-27 11:06:10 +00:00
backend for sessionsetup call - this takes all 3 variants of the call
first public release of samba4 code (This used to be commit b0510b5428b3461aeb9bbe3cc95f62fc73e2b97f)
2003-08-13 01:53:07 +00:00
*/
r1280: rename struct request_context to smbsrv_request metze (This used to be commit a85d2db5826a84b812ea5162a11f54edd25f74e3)
2004-06-28 08:39:00 +00:00
NTSTATUS sesssetup_backend(struct smbsrv_request *req,
first public release of samba4 code (This used to be commit b0510b5428b3461aeb9bbe3cc95f62fc73e2b97f)
2003-08-13 01:53:07 +00:00
union smb_sesssetup *sess)
{
r1507: fixed the handling of SMB chaining with the new server structure. You must think carefully about packet chaining when dealing with any authentication or SMB parsing issues. The particular problem here was that a chained tconX didn't get the req->session setup after an initial sesstion setup call, so the tconx used a bogus VUID. (This used to be commit 6f2a335cd623211071b01d982d4e7c69b49a5602)
2004-07-15 02:11:03 +00:00
NTSTATUS status = NT_STATUS_INVALID_LEVEL;
r4777: added a smb_composite_sesssetup() async composite function. This encapsulates all the different session setup methods, including the multi-pass spnego code. I have hooked this into all the places that previously used the RAW_SESSSETUP_GENERIC method, and have removed the old RAW_SESSSETUP_GENERIC code from clisession.c and clitree.c. A nice side effect is that these two modules are now very simple again, back to being "raw" session setup handling, which was what was originally intended. I have also used this to replace the session setup code in the smb_composite_connect() code, and used that to build a very simple replacement for smbcli_tree_full_connection(). As a result, smbclient, smbtorture and all our other SMB connection code now goes via these composite async functions. That should give them a good workout! (This used to be commit 080d0518bc7d6fd4bc3ef783e7d4d2e3275d0799)
2005-01-16 11:15:08 +00:00
switch (sess->old.level) {
first public release of samba4 code (This used to be commit b0510b5428b3461aeb9bbe3cc95f62fc73e2b97f)
2003-08-13 01:53:07 +00:00
case RAW_SESSSETUP_OLD:
r1507: fixed the handling of SMB chaining with the new server structure. You must think carefully about packet chaining when dealing with any authentication or SMB parsing issues. The particular problem here was that a chained tconX didn't get the req->session setup after an initial sesstion setup call, so the tconx used a bogus VUID. (This used to be commit 6f2a335cd623211071b01d982d4e7c69b49a5602)
2004-07-15 02:11:03 +00:00
status = sesssetup_old(req, sess);
break;
first public release of samba4 code (This used to be commit b0510b5428b3461aeb9bbe3cc95f62fc73e2b97f)
2003-08-13 01:53:07 +00:00
case RAW_SESSSETUP_NT1:
r1507: fixed the handling of SMB chaining with the new server structure. You must think carefully about packet chaining when dealing with any authentication or SMB parsing issues. The particular problem here was that a chained tconX didn't get the req->session setup after an initial sesstion setup call, so the tconx used a bogus VUID. (This used to be commit 6f2a335cd623211071b01d982d4e7c69b49a5602)
2004-07-15 02:11:03 +00:00
status = sesssetup_nt1(req, sess);
break;
first public release of samba4 code (This used to be commit b0510b5428b3461aeb9bbe3cc95f62fc73e2b97f)
2003-08-13 01:53:07 +00:00
case RAW_SESSSETUP_SPNEGO:
r1507: fixed the handling of SMB chaining with the new server structure. You must think carefully about packet chaining when dealing with any authentication or SMB parsing issues. The particular problem here was that a chained tconX didn't get the req->session setup after an initial sesstion setup call, so the tconx used a bogus VUID. (This used to be commit 6f2a335cd623211071b01d982d4e7c69b49a5602)
2004-07-15 02:11:03 +00:00
status = sesssetup_spnego(req, sess);
break;
first public release of samba4 code (This used to be commit b0510b5428b3461aeb9bbe3cc95f62fc73e2b97f)
2003-08-13 01:53:07 +00:00
}
r1507: fixed the handling of SMB chaining with the new server structure. You must think carefully about packet chaining when dealing with any authentication or SMB parsing issues. The particular problem here was that a chained tconX didn't get the req->session setup after an initial sesstion setup call, so the tconx used a bogus VUID. (This used to be commit 6f2a335cd623211071b01d982d4e7c69b49a5602)
2004-07-15 02:11:03 +00:00
if (NT_STATUS_IS_OK(status)) {
req->smb_conn->negotiate.done_sesssetup = True;
}
first public release of samba4 code (This used to be commit b0510b5428b3461aeb9bbe3cc95f62fc73e2b97f)
2003-08-13 01:53:07 +00:00
r1507: fixed the handling of SMB chaining with the new server structure. You must think carefully about packet chaining when dealing with any authentication or SMB parsing issues. The particular problem here was that a chained tconX didn't get the req->session setup after an initial sesstion setup call, so the tconx used a bogus VUID. (This used to be commit 6f2a335cd623211071b01d982d4e7c69b49a5602)
2004-07-15 02:11:03 +00:00
return status;
first public release of samba4 code (This used to be commit b0510b5428b3461aeb9bbe3cc95f62fc73e2b97f)
2003-08-13 01:53:07 +00:00
}
Copy Permalink