summaryrefslogtreecommitdiff
path: root/source4
AgeCommit message (Collapse)AuthorFilesLines
2013-08-23libsmb2: Fix opening the rootdirectory, part 2Volker Lendecke1-1/+1
smb2_push_o16s16_blob is wrong for the blob.data==NULL case. It does not do the same magic that the rest of the routine does with regards to padding_fix. padding_fix is wrong in its own respect, with a 0-length blob we end up with a negative padding fix. It's wrong, but it seems to work. Why am I doing this? I want to make smb2.getinfo work against w2k12. smb2_util_roothandle() always gives NT_STATUS_INVALID_PARAMETER without this and the preceding fix. Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
2013-08-23libsmb2: Fix opening the rootdirectory, part 1Volker Lendecke1-0/+15
[MS-SMB2], 2.2.13 says: In the request, the Buffer field MUST be at least one byte in length. Implement that for the 0-length filename without create blobs. Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
2013-08-23torture: Remove an unused variableVolker Lendecke1-3/+0
Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
2013-08-19registry4: Fix CID 1034911 Dereference before null checkVolker Lendecke1-1/+1
curbegin is always != NULL here (curend + 1) and is dereferenced by strchr. Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
2013-08-19samdb: Fix CID 1034910 Dereference before null checkVolker Lendecke1-3/+3
strncmp("tdb://", sam_name, 6) dereferences sam_name. Check for NULL before that. Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
2013-08-19samdb: Fix CID 1034910 Dereference before null checkVolker Lendecke1-3/+3
strncmp("tdb://", sam_name, 6) dereferences sam_name. Check for NULL before that. Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
2013-08-19samdb: Fix CID 1034910 Dereference before null checkVolker Lendecke1-3/+3
strncmp("tdb://", secrets_ldb, 6) dereferences secrets_ldb. Check for NULL before that. Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
2013-08-19registry4: Fix CID 1034911 Dereference before null checkVolker Lendecke1-1/+1
curbegin is always != NULL here (curend + 1) and is dereferenced by strchr. Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
2013-08-15torture: support Windows 2k8 response for compress_invalid_bufDavid Disseldorp1-4/+6
Windows Server 2012 returns NT_STATUS_INVALID_USER_BUFFER, Windows Server 2008r2 returns NT_STATUS_INVALID_PARAMETER. Don't fail the test if either status is returned. Signed-off-by: David Disseldorp <ddiss@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
2013-08-15torture: add more [no-]compress-on-open ioctl testsDavid Disseldorp1-1/+144
compress_create_with_attr: Specifies the FILE_ATTRIBUTE_COMPRESSED attribute at create time, then checks the created file. compress_inherit_disable: Creates under a compressed directory, a file with the NTCREATEX_OPTIONS_NO_COMPRESSION option. Then checks that the newly created file doesn't inherit the parent compression state. Signed-off-by: David Disseldorp <ddiss@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
2013-08-15torture: add file_attribs arg to file create helperDavid Disseldorp1-24/+23
Allows for the testing of file creation with FILE_ATTRIBUTE_COMPRESSED. Signed-off-by: David Disseldorp <ddiss@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
2013-08-15torture: add compressed file attribute ioctl testDavid Disseldorp1-0/+53
After marking a file for compression via FSCTL_SET_COMPRESSION, the FILE_ATTRIBUTE_COMPRESSED flag should be present. Test for this. Signed-off-by: David Disseldorp <ddiss@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
2013-08-15torture: check for filesystem compression capabilityDavid Disseldorp1-18/+59
Only run the compression tests if the compression capability is returned by the server in response to an SMB2_QUERY_INFO(FS attribute info) request. Signed-off-by: David Disseldorp <ddiss@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
2013-08-15torture: extend FSCTL_[GET/SET]_COMPRESSION testsDavid Disseldorp1-39/+234
Check for inheritance of compression attributes from parent directories. Also, test error handling for invalid requests. Signed-off-by: David Disseldorp <ddiss@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
2013-08-15libcli/smb: pass max_dyn_len to smb2cli_req_send()Stefan Metzmacher1-3/+6
This way we can calculate the correct credit charge for requests with large output buffers. Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
2013-08-12libcli: Fix improper use of tevent_req_simple_recv_ntstatusVolker Lendecke1-6/+10
tevent_req_simple_recv_ntstatus is just for the simple return without anything to do after it. Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
2013-08-12lsa4: Fix an set but unused variable warningVolker Lendecke1-4/+4
Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
2013-08-12lsa4: Remove an unused variableVolker Lendecke1-3/+0
Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
2013-08-12s4:wrepl_out_helpers.c: avoid talloc_reference() in most casesStefan Metzmacher1-3/+18
Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
2013-08-12s4:torture/rpc: s/getgroups/getgr to avoid compiler warningsStefan Metzmacher1-5/+5
Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
2013-08-12s4:nbt_server: avoid talloc_reference()Stefan Metzmacher3-5/+6
Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
2013-08-12s4:torture/rpc/samsync: avoid talloc_reference()Stefan Metzmacher1-14/+13
Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
2013-08-12s4:libcli: avoid talloc_reference() in finddcs_nbt_send()Stefan Metzmacher1-1/+1
Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
2013-08-12s4:librpc: avoid talloc_reference() in dcerpc_epm_map_binding_send()Stefan Metzmacher3-12/+35
Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
2013-08-12waf: replace dependency to libintl with samba_intlChristian Ambach1-2/+2
Signed-off-by: Christian Ambach <ambi@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org> Autobuild-User(master): Christian Ambach <ambi@samba.org> Autobuild-Date(master): Mon Aug 12 00:46:34 CEST 2013 on sn-devel-104
2013-08-09waf: consolidate libintl related checksChristian Ambach1-3/+1
consolidate the dealing with functions from libintl and the handling of checking if libiconv is required or not to a common place in lib/replace also add a new samba_intl subsystem that has dependencies on the appropriate set of libraries (libintl, libintl+libiconv or none) that can be used as a general dependency by code that depends on the internationalization libraries Signed-off-by: Christian Ambach <ambi@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
2013-08-10auth/gensec: treat struct gensec_security_ops as const if possible.Stefan Metzmacher1-2/+2
Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
2013-08-10auth/gensec: introduce gensec_internal.hStefan Metzmacher9-0/+9
We should treat most gensec related structures private. It's a long way, but this is a start. Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
2013-08-10s4:gensec/schannel: only require librpc/gen_ndr/dcerpc.hStefan Metzmacher1-1/+1
We just need DCERPC_AUTH_TYPE_SCHANNEL Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
2013-08-10s4:gensec/schannel: there's no point in having schannel_session_key()Stefan Metzmacher1-8/+0
gensec_session_key() will return NT_STATUS_NO_USER_SESSION_KEY before calling schannel_session_key(), as we don't provide GENSEC_FEATURE_SESSION_KEY. Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
2013-08-10s4:gensec/schannel: GENSEC_FEATURE_ASYNC_REPLIES is not supportedStefan Metzmacher1-3/+0
There's a sequence number attached to the connection, which needs to be incremented with each message... Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
2013-08-10s4:gensec/schannel: use the correct computer_name from ↵Stefan Metzmacher1-3/+3
netlogon_creds_CredentialState We need to use the same computer_name we used in the netr_Authenticate3 request. Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
2013-08-10s4:gensec/schannel: simplify the code by using netsec_create_state()Stefan Metzmacher1-68/+30
Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
2013-08-10s4:gensec/schannel: remove unused dcerpc_schannel_creds()Stefan Metzmacher2-49/+0
Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
2013-08-10s4:torture: avoid usage of dcerpc_schannel_creds()Stefan Metzmacher4-17/+11
We use cli_credentials_get_netlogon_creds() which returns the same value. dcerpc_schannel_creds() is a layer violation. Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
2013-08-10s4:libnet: avoid usage of dcerpc_schannel_creds()Stefan Metzmacher1-4/+3
We use cli_credentials_get_netlogon_creds() which returns the same value. dcerpc_schannel_creds() is a layer violation. Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
2013-08-09torture: add smb2 FSCTL_[GET/SET]_COMPRESSION testDavid Disseldorp1-0/+84
This test simply creates a file and checks the compression state before and after FSCTL_SET_COMPRESSION(COMPRESSION_FORMAT_DEFAULT). The test expects the compression state to be COMPRESSION_FORMAT_LZNT1 after set, conforming to Windows Server behaviour. If the server responds to the first FSCTL_GET_COMPRESSION request with NT_STATUS_NOT_SUPPORTED or NT_STATUS_INVALID_DEVICE_REQUEST, then the test is skipped. This allows it to run during selftest. Signed-off-by: David Disseldorp <ddiss@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org> Autobuild-User(master): Jeremy Allison <jra@samba.org> Autobuild-Date(master): Fri Aug 9 22:03:39 CEST 2013 on sn-devel-104
2013-08-09torture: split out ioctl test file creation helperDavid Disseldorp1-56/+43
Signed-off-by: David Disseldorp <ddiss@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
2013-08-07drs-cracknames: Add some debugs in the torture to know better which test has ↵Matthieu Patou1-0/+4
failed Signed-off-by: Matthieu Patou <mat@matws.net> Reviewed-By: Andrew Bartlett <abarlett@samba.org> Autobuild-User(master): Matthieu Patou <mat@samba.org> Autobuild-Date(master): Wed Aug 7 08:10:58 CEST 2013 on sn-devel-104
2013-08-06drs-cracksname: fix problems that prevented to pass our torture testsMatthieu Patou1-6/+21
Some of the problems where also reported by Microsoft testing tools Signed-off-by: Matthieu Patou <mat@matws.net> Reviewed-by: Andrew Bartlett <abarlett@samba.org>
2013-08-06drs-crackname: Fix error code so that we have the same as windowsMatthieu Patou1-0/+1
Signed-off-by: Matthieu Patou <mat@matws.net> Reviewed-by: Andrew Bartlett <abarlett@samba.org>
2013-08-06drs-cracknames: When cracking NT4 names we should just look at netbios for ↵Matthieu Patou1-2/+1
the match Looking at dnsRoot will yield a result for domain.tld\username when it shouldn't work. Signed-off-by: Matthieu Patou <mat@matws.net> Reviewed-by: Andrew Bartlett <abarlett@samba.org>
2013-08-06drs-crackname: Fix cracknames for the format UNKNOWN when the data is ↵Matthieu Patou1-1/+4
actually a GUID The cannonical crackname expect a "/" or it returns DRSUAPI_DS_NAME_STATUS_RESOLVE_ERROR, when doing UNKNOWN format it's not an error to not have a "/" in the name to crack it's just a sign the name is not a cannonical one. Signed-off-by: Matthieu Patou <mat@matws.net> Reviewed-by: Andrew Bartlett <abarlett@samba.org>
2013-08-06drs-cracknames: Reorganise the cracknames list so that similar format names ↵Matthieu Patou1-6/+6
are group together It makes easier when reviewing failed test case in DRSR testsuite Signed-off-by: Matthieu Patou <mat@matws.net> Reviewed-by: Andrew Bartlett <abarlett@samba.org>
2013-08-06Add Notes related to DRSUAPIMatthieu Patou1-0/+1
Signed-off-by: Matthieu Patou <mat@matws.net> Reviewed-by: Andrew Bartlett <abarlett@samba.org>
2013-08-06s4-netlogon: honnor DS_RETURN_DNS_NAME flagMatthieu Patou1-0/+9
Reviewed-By: Andrew Bartlett <abarlett@samba.org>
2013-08-06s4-netlogon: do not add \\ it has already be done in the ↵Matthieu Patou1-1/+3
fill_netlogon_samlogon_response Reviewed-By: Andrew Bartlett <abarlett@samba.org>
2013-08-06torture: Quiet a warning about set but not used variableMatthieu Patou1-8/+7
Signed-off-by: Matthieu Patou <mat@matws.net> Reviewed-By: Andrew Bartlett <abarlett@samba.org>
2013-08-06torture-drsuapi: Make the name of the dc variableMatthieu Patou1-3/+5
In case some tests fails or if the removal takes sometime to replicate to all the DCs Reviewed-By: Andrew Bartlett <abarlett@samba.org>
2013-08-05Fix bug #10010 - Missing integer wrap protection in EA list reading can ↵Jeremy Allison1-2/+5
cause server to loop with DOS. Fix client-side parsing also. Found by David Disseldorp <ddiss@suse.de> Signed-off-by: Jeremy Allison <jra@samba.org> Autobuild-User(master): Karolin Seeger <kseeger@samba.org> Autobuild-Date(master): Mon Aug 5 14:39:04 CEST 2013 on sn-devel-104