summaryrefslogtreecommitdiff
path: root/source4/libcli/smb2
AgeCommit message (Collapse)AuthorFilesLines
2013-01-16smb2_ioctl: remove ioctl error response assumptionsDavid Disseldorp1-3/+34
MS-SMB2 3.3.4.4 documents cases where a ntstatus indicating an error should not be considered a failure. In such a case the output data buffer should be sent to the client rather than an error response packet. Add a new fsctl copy_chunk test to confirm field limits are sent back in response to an oversize chunk request. Reviewed by: Jeremy Allison <jra@samba.org>
2012-09-19s4:libci: add a SMB2_CREATE_APP_INSTANCE_ID blob to the request if the ↵Michael Adam1-0/+25
in.app_instance_id is present
2012-08-17s4:libcli/smb2: reset trsnport->compound.related when a compound chain is ↵Stefan Metzmacher1-0/+1
finished metze
2012-08-06s4:libcli/smb2/write correct error checkingChristian Ambach1-1/+1
Server might return STATUS_BUFFER_OVERFLOW, which is not caught by NT_STATUS_IS_ERR Autobuild-User(master): Christian Ambach <ambi@samba.org> Autobuild-Date(master): Mon Aug 6 20:01:01 CEST 2012 on sn-devel-104
2012-07-25libcli/smb: pass smbXcli_session to smb2cli_tcon_set_values()Stefan Metzmacher1-0/+1
metze
2012-07-25s4:libcli/smb2: remove unused smb2_session->pidStefan Metzmacher3-4/+0
metze
2012-07-25s4:libcli/smb2: remove unused variableStefan Metzmacher1-2/+0
metze
2012-07-25libcli/smb: there's no PID field in the SMB2/3 header anymoreStefan Metzmacher1-4/+0
It's a reserved field... metze
2012-07-25libcli/smb: pass smbXcli_tcon to smb2cli_req_create/send()Stefan Metzmacher1-4/+8
metze
2012-07-25s4:libcli/smb2: remove unused elements from smb2_treeStefan Metzmacher3-6/+0
metze
2012-07-24s4:libcli/smb2: setup a smbXcli_tcon for each smb2_treeStefan Metzmacher3-0/+16
metze
2012-05-06s4:libcli/smb2: use PROTOCOL_LATESTStefan Metzmacher1-1/+1
metze
2012-04-16s4:libcli/smb2: fix anonymous session setups against windows serversStefan Metzmacher2-1/+16
Windows server doesn't set the SMB2_SESSION_FLAG_IS_GUEST nor SMB2_SESSION_FLAG_IS_NULL flag. This fix makes sure we don't try to verify a signature on the final session setup response. metze Autobuild-User: Stefan Metzmacher <metze@samba.org> Autobuild-Date: Mon Apr 16 14:44:46 CEST 2012 on sn-devel-104
2012-04-16s4:libcli/smb2: remove unused dependency to LIBPACKETStefan Metzmacher2-2/+1
metze
2012-03-25LIBCLI_SMB2: Depend on cli_smb_common rather than libsmb.Jelmer Vernooij1-1/+1
The latter pulls in a fair number of other libraries, including popt and the Samba registry library.
2012-03-15libsmbclient-raw: Install smb_composite.h.Jelmer Vernooij1-1/+2
Autobuild-User: Jelmer Vernooij <jelmer@samba.org> Autobuild-Date: Thu Mar 15 02:57:59 CET 2012 on sn-devel-104
2012-03-07s4:libcli/smb2: allow smb2_session_setup_spnego to handle reauthGregor Beck1-0/+12
Pair-Programmed-With: Stefan Metzmacher <metze@samba.org> Signed-off-by: Stefan Metzmacher <metze@samba.org>
2012-03-03s4:libcli:smb2: set SMB2_CAP_ALL in the negprotMichael Adam1-1/+5
2012-03-03s4:libcli:smb2: store the share capabilites in the smb2_tree objectMichael Adam2-0/+2
2012-03-03smbXcli: add the possiblilty to negotiate client capabilites in smb >= 2.2Michael Adam1-1/+2
Pair-Programmed-With: Stefan Metzmacher <metze@samba.org>
2012-03-03s4:libcli:smb2: allow max protocol 0x0224Michael Adam1-1/+1
2012-03-03s4:libcli:smb2: add support for parsing the durable handle v2 response in ↵Michael Adam1-0/+19
smb2_create_recv()
2012-03-03s4:libcli:smb2: add support durable handle reconnect v2 blob in smb2_create_sendMichael Adam1-0/+27
2012-03-03s4:libcli:smb2: add support durable handle request v2 blob in smb2_create_sendMichael Adam1-0/+28
2012-02-28s4:libcli:smb2: add a smb2_connect_ext variant of smb2_connect that takes a ↵Michael Adam1-12/+36
previous_session argument
2012-02-28s4:libcli:smb2: add a previous session argument to smb2_connect_send()Michael Adam1-1/+5
2012-02-28s4:libcli:smb2: add a previous session argument to smb2_session_setup_spnego()Michael Adam2-9/+16
So that we can do a session reconnect.
2012-02-28s3:libsmb:smb2: pass previous session ID to session setup, not previous sessionMichael Adam1-2/+2
We only need the session, and under some circumstances, it might complicate things for the caller to have to cope with the whole structure (talloc...).
2012-02-27s4:libcli/smb2: remove unused 'session_key' from struct smb2_sessionStefan Metzmacher2-4/+5
metze
2012-02-27libcli/smb/smbXcli: maintain smb2 channel_signing_key separate from the ↵Stefan Metzmacher1-3/+3
signing_key The signing_key is fix across all channels and is used for session setups on a channel binding. Note: - the last session setup response is signed with the new channel signing key. - the reauth session setups are signed with the channel signing key. It's also not needed to remember the main session key. metze
2012-02-23s4:smbcli:smb2: add a random GUID to the transport connection in ↵Michael Adam1-1/+5
smb2_transport_init() This GUID is used in the smb2 negprot when max protocol is bigger than 0x0202. According to section 2.2.3 of the MS-SMB2 document, the Client GUID filed in the SMB2 negotiate request must be filled with a (non-zero) GUID if there are other dialects than 0x0202 in the dialects field. http://msdn.microsoft.com/en-us/library/cc246543%28v=prot.13%29.aspx Apart from corresponding to the docs, this change makes some of our durable-open tests (e.g reopen2 and open-oplock) _not_ hang when running against windows 8 preview (which might be still buggy). Pair-Programmed-With: Gregor Beck <gbeck@sernet.de> Autobuild-User: Michael Adam <obnox@samba.org> Autobuild-Date: Thu Feb 23 03:23:57 CET 2012 on sn-devel-104
2012-02-03s4:libcli:smb2: don't allocate the smb2_tree struct earlier than necessary ↵Michael Adam1-5/+5
in smb2_connect() Pair-Programmed-With: Stefan Metzmacher <metze@samba.org>
2012-02-03s4:libcli/torture:smb2: change smb2_tree_connect to take a smb2_session ↵Michael Adam2-6/+6
instead of smb2_tree arguement. Only the tree->session is needed. This way it is less confusing. Pair-Programmed-With: Stefan Metzmacher <metze@samba.org>
2011-12-09s4:libcli/smb2: mark a request as error if we can't ship itStefan Metzmacher1-0/+2
metze Autobuild-User: Stefan Metzmacher <metze@samba.org> Autobuild-Date: Fri Dec 9 15:13:11 CET 2011 on sn-devel-104
2011-11-30s4:libcli: do the nbss session request within smbcli_sock_connect_*()Stefan Metzmacher1-39/+17
metze
2011-11-29s4:libcli/smb2: make sure only one idle event runs at a timeStefan Metzmacher2-9/+17
metze
2011-11-24s4:libcli/smb2: use talloc_zero() in smb2_request_init()Stefan Metzmacher1-16/+3
metze
2011-11-24s4:libcli/smb2: fix compiler warning in smb2_push_o16s16_string()Stefan Metzmacher1-1/+3
metze
2011-11-24s4:libcli/smb2: implement on top of smbXcli_conn/reqStefan Metzmacher11-863/+422
metze
2011-11-03s4:libcli/smb2: s/SMB_SIGNING_SUPPORTED/SMB_SIGNING_IF_REQUIRED/Stefan Metzmacher1-2/+2
metze
2011-11-03s4:libcli/smb2: remove unused SMB_SIGNING_AUTO handlingStefan Metzmacher1-2/+0
metze
2011-11-03s4:libcli/smb2: SMB_SIGNING_DEFAULT matches SMB_SIGNING_SUPPORTED on the ↵Stefan Metzmacher1-0/+2
client for now metze
2011-11-03s4:libcli/smb2: let SMB_SIGNING_AUTO behave like SMB_SIGNING_SUPPORTEDStefan Metzmacher1-7/+1
This matches the smb1 behavior. metze
2011-10-30s4:libcli: in smb2_create_recv, map SMB2_CREATE_DURABLE_HANDLE_RESPONSE ↵Michael Adam1-0/+7
buffer smb2_create.out.durable_open
2011-10-27replace: Add don't include unistd.h directly and add uid_wrapper.Andreas Schneider1-2/+1
2011-10-19s4:libcli/smb2: make use of _smb_setlen_tcp()Stefan Metzmacher1-2/+2
metze
2011-10-18gensec: move event context from gensec_*_init() to gensec_update()Andrew Bartlett1-1/+2
This avoids keeping the event context around on a the gensec_security context structure long term. In the Samba3 server, the event context we either supply is a NULL pointer as no server-side modules currently use the event context. Andrew Bartlett Signed-off-by: Stefan Metzmacher <metze@samba.org>
2011-09-28s4:libcli/smb2: ignore SMB2_OP_CANCEL responsesStefan Metzmacher1-0/+10
If there're a problem with signing or the session doesn't exists any more the server responses with a failure, instead of not sending a response. For now we ignore the reponse, as there's not much we could do with it and it's not likely that we generate bad requests, which trigger that behavior, except for testing. metze
2011-09-28s4:libcli/smb2: correctly sign SMB2_OP_CANCEL request if they belong to a ↵Stefan Metzmacher1-0/+1
session metze
2011-09-28s4:libcli/smb2: don't try to check the signing if we got ↵Stefan Metzmacher1-1/+2
NT_STATUS_USER_SESSION_DELETED metze