summaryrefslogtreecommitdiff
path: root/source3/smbd/process.c
AgeCommit message (Collapse)AuthorFilesLines
2012-03-10s3: Add smb_request_doneVolker Lendecke1-58/+95
This is used to enable async chained command sequences. A synchronous reply_xxx command does not need to take are anymore about and_x chaining. The async commands (pipe r/w at this moment) must do so however. When finished, they must inform the main chain engine that they are finished with a smb_request_done call. Autobuild-User: Volker Lendecke <vl@samba.org> Autobuild-Date: Sat Mar 10 17:14:05 CET 2012 on sn-devel-104
2012-03-10s3: Remove unused smb_request->doneVolker Lendecke1-6/+0
2012-03-10s3: Remove unused smb_request->chain_outbufVolker Lendecke1-1/+0
2012-03-10s3: Remove "req_wct_ofs()"Volker Lendecke1-19/+0
This is fixed up in construct_reply_chain
2012-03-10s3: Fix the read&x offset within a chainVolker Lendecke1-0/+32
2012-03-10s3: Remove chain_replyVolker Lendecke1-270/+0
<insert your favourite tombstone ascii art here>
2012-03-10s3: Replace chain_replyVolker Lendecke1-2/+135
This is a new implementation of our andx handling code. The old code was quite involved in that it was called from within the reply_ handlers. This leads to pretty complex faking of smb_request structures to give them the same environment, independent of whether they are called directly or from within chain_reply. chain_reply needs to go because it blocks really async handling of chained requests.
2012-03-10s3: Add a new set of andx chain handling routinesVolker Lendecke1-0/+236
This is in preparation of getting rid of chain_reply.
2012-03-06s3:smbd: keep 'num_users' and 'users' directly under smbd_server_connectionStefan Metzmacher1-6/+1
The plan is to have users_struct as some kind of low level abstraction for a smb1/smb2 session, that can be used by SMB_VFS modules. metze
2012-03-06s3: Move a talloc_strdup out of the main code pathVolker Lendecke1-15/+31
This is only used for AS_GUEST requests Autobuild-User: Volker Lendecke <vl@samba.org> Autobuild-Date: Tue Mar 6 14:29:50 CET 2012 on sn-devel-104
2012-03-05s3: Move the drain_socket on error to reply_write_and_XVolker Lendecke1-9/+0
That's the only case where this can happen, so we should not clutter the main code path.
2012-03-05s3: Remove "size" param from switch_messageVolker Lendecke1-3/+3
Autobuild-User: Volker Lendecke <vl@samba.org> Autobuild-Date: Mon Mar 5 15:13:49 CET 2012 on sn-devel-104
2012-03-05s3: Remove "size" param from smb_dumpVolker Lendecke1-4/+5
2012-03-04s3-auth: Remove security=share (depricated since 3.6).Andrew Bartlett1-6/+1
This patch removes security=share, which Samba implemented by matching the per-share password provided by the client in the Tree Connect with a selection of usernames supplied by the client, the smb.conf or guessed from the environment. The rationale for the removal is that for the bulk of security=share users, we just we need a very simple way to run a 'trust the network' Samba server, where users mark shares as guest ok. This is still supported, and the smb.conf options are documented at https://wiki.samba.org/index.php/Public_Samba_Server At the same time, this closes the door on one of the most arcane areas of Samba authentication. Naturally, full user-name/password authentication remain available in security=user and above. This includes documentation updates for username and only user, which now only do a small amount of what they used to do. Andrew Bartlett -------------- / \ / REST \ / IN \ / PEACE \ / \ | SEC_SHARE | | security=share | | | | | | 5 March | | | | 2012 | *| * * * | * _________)/\\_//(\/(/\)/\//\/\///|_)_______
2012-02-29Add the implementation of check_reduced_name_with_privilege(). Now to plumb intoJeremy Allison1-0/+1
SMB1 requests.
2012-02-29s3: Fix a const warningVolker Lendecke1-1/+1
2012-02-27s3-param: Align lp_{max,min}protocol with lib/param namesAndrew Bartlett1-1/+1
This adds an alisas to ensure that both our loadparm systems know all the names. I would like to move to the 'server ..' name as canonical, and this will be raised on the list. Andrew Bartlett
2012-02-24s3: Simplify smb_splice_chainVolker Lendecke1-16/+13
No code change, just remove a {} block left over as the else branch from the previous commit Autobuild-User: Volker Lendecke <vl@samba.org> Autobuild-Date: Fri Feb 24 18:07:48 CET 2012 on sn-devel-104
2012-02-24s3: Simplify smb_splice_chainVolker Lendecke1-12/+2
first_request won't be true anymore, we always splice fully existing records in smbd
2012-02-24s3: Simplify smb_splice_chainVolker Lendecke1-13/+2
With the last commit, bytes_padding is not set anymore
2012-02-24s3: Simplify smb_splice_chainVolker Lendecke1-7/+1
bytes_alignment used to be used in libsmb, which uses different code now
2012-02-24s3: Simplify smb_splice_chainVolker Lendecke1-25/+11
We use it in smbd/process.c only now. This only splices in finished buffers, both callers used the same arguments. Pull that into the routine itself.
2012-02-24s3: Turn some SMB_ASSERTS into proper returnVolker Lendecke1-2/+6
We deal with the error properly further up
2012-02-24s3: Fix a DEBUG msgVolker Lendecke1-1/+1
2012-02-23s3: smb_request->vwv can be constVolker Lendecke1-1/+1
Autobuild-User: Volker Lendecke <vl@samba.org> Autobuild-Date: Thu Feb 23 12:37:23 CET 2012 on sn-devel-104
2012-02-18Fix a bunch of "unused variable" warnings.Jeremy Allison1-3/+0
Autobuild-User: Jeremy Allison <jra@samba.org> Autobuild-Date: Sat Feb 18 06:22:40 CET 2012 on sn-devel-104
2012-02-17Replace smbd_server_connection_loop_once() with tevent_loop_once() directly.Jeremy Allison1-63/+6
We no longer need to call poll() directly inside smbd ! Autobuild-User: Jeremy Allison <jra@samba.org> Autobuild-Date: Fri Feb 17 02:49:13 CET 2012 on sn-devel-104
2012-02-16lib/util: Remove sys_poll as it is no longer neededAndrew Bartlett1-1/+1
sys_poll() is only needed if the signal pipe is set up and used, but as no signal handler ever writes to the pipe, this can all be removed. signal based events are now handled via tevent. Andrew Bartlett Signed-off-by: Jeremy Allison <jra@samba.org>
2012-01-05s3: Move basic SMB checking to a much earlier pointVolker Lendecke1-19/+19
2012-01-05s3: Add a suicide mode to smbdVolker Lendecke1-0/+11
To test our cleanup code paths properly, we need a way to make smbd exit hard without cleaning up
2012-01-04s3: Check for the packet size before accessing itVolker Lendecke1-2/+1
Autobuild-User: Jeremy Allison <jra@samba.org> Autobuild-Date: Wed Jan 4 15:16:20 CET 2012 on sn-devel-104
2011-12-15s3:smbd: pass smbd_server_connection and a snumused function pointer to ↵Stefan Metzmacher1-4/+4
reload_services() metze
2011-12-15s3:smbd: split smb_conf_updated into parent and child versionsStefan Metzmacher1-0/+21
metze
2011-12-15s3:smbd: split ID_CACHE_* message handling into parent and child partsStefan Metzmacher1-0/+111
metze
2011-12-13s3:smbd: pass smbd_server_connection to smbd_setup_sig_hup_handler()Stefan Metzmacher1-10/+11
metze
2011-12-13s3:smbd: pass smbd_server_connection to smbd_setup_sig_term_handler()Stefan Metzmacher1-4/+4
metze
2011-12-13s3:smbd/oplock: pass smbd_server_connection to init_oplocks()Stefan Metzmacher1-1/+1
metze
2011-12-13s3:smbd/open: pass smbd_server_connection as private_data to ↵Stefan Metzmacher1-1/+1
msg_file_was_renamed() metze
2011-12-13s3:smbd: register MSG_SMB_FILE_RENAME after the forkStefan Metzmacher1-0/+2
The parent smbd doesn't need to handle this, as it doesn't have any file handles open. metze
2011-12-13s3:smbd/conn_msg: pass smbd_server_connection as private_data to ↵Stefan Metzmacher1-1/+1
msg_force_tdis() metze
2011-12-12s3: Remove a bunch of calls to procid_self()Volker Lendecke1-1/+1
All callers to messaging_[re]init only used procid_self()
2011-12-12s3:smbd/clode: pass smbd_server_connection as private_data to msg_close_file()Stefan Metzmacher1-1/+1
metze
2011-12-12s3:smbd/process: avoid using server_event_context() for the forked echo handlerStefan Metzmacher1-3/+3
metze
2011-12-12s3:smbd/process: avoid using server_event_context() for ↵Stefan Metzmacher1-10/+10
smbd_deferred_open_timer events metze
2011-12-12s3:smbd: pass down smbd_server_connection via smbd_echo_stateStefan Metzmacher1-4/+5
metze
2011-12-12s3:smbd: remove references to the global smbd_server_connStefan Metzmacher1-2/+2
metze
2011-12-12s3:smbd: make struct pending_message_list privateStefan Metzmacher1-0/+14
metze
2011-12-12s3:smbd: remember the smbd_server_connection on pending_message_listStefan Metzmacher1-4/+6
metze
2011-12-12s3:smbd: call sub_set_socket_ids() in smbd_process() againStefan Metzmacher1-0/+18
This got lost in commit b2511a280aa9449123376fd3cbb495dcd1a87dee. metze Autobuild-User: Stefan Metzmacher <metze@samba.org> Autobuild-Date: Mon Dec 12 10:23:44 CET 2011 on sn-devel-104
2011-11-10s3: Avoid a race with the async echo handlerVolker Lendecke1-33/+28
We can not read from the echo handler socket when we have the main socket locked. This leads to the echo responder to lock up sitting in the fcntl lock while the parent wants to read the remainder of a large packet.