nivanova/samba-autobuild/.git
8 years agoauth/ntlmssp: implement GENSEC_FEATURE_LDAP_STYLE
Stefan Metzmacher [Wed, 9 Dec 2015 13:48:14 +0000 (14:48 +0100)]
auth/ntlmssp: implement GENSEC_FEATURE_LDAP_STYLE

We need to handle NTLMSSP_NEGOTIATE_SIGN as
NTLMSSP_NEGOTIATE_SEAL if GENSEC_FEATURE_LDAP_STYLE
is requested.

This works arround a bug in Windows, which allow signed only
messages using NTLMSSP and LDAP.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
8 years agoauth/gensec: add GENSEC_FEATURE_LDAP_STYLE define
Stefan Metzmacher [Wed, 9 Dec 2015 13:48:14 +0000 (14:48 +0100)]
auth/gensec: add GENSEC_FEATURE_LDAP_STYLE define

This will be used for LDAP connections and may trigger
backend specific behaviour.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
8 years agoauth/ntlmssp: use ndr_push_AV_PAIR_LIST in gensec_ntlmssp_server_negotiate().
Günther Deschner [Tue, 18 Aug 2009 22:40:12 +0000 (00:40 +0200)]
auth/ntlmssp: use ndr_push_AV_PAIR_LIST in gensec_ntlmssp_server_negotiate().

Pair-Programmed-With: Stefan Metzmacher <metze@samba.org>

Signed-off-by: Günther Deschner <gd@samba.org>
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agolibrpc/ndr: add ndr_ntlmssp_find_av() helper function
Stefan Metzmacher [Tue, 24 Nov 2015 14:40:29 +0000 (15:40 +0100)]
librpc/ndr: add ndr_ntlmssp_find_av() helper function

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
8 years agontlmssp.idl: make AV_PAIR_LIST public
Stefan Metzmacher [Thu, 19 Nov 2015 14:38:02 +0000 (15:38 +0100)]
ntlmssp.idl: make AV_PAIR_LIST public

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
8 years agontlmssp.idl: MsAvRestrictions is MsvAvSingleHost now
Stefan Metzmacher [Mon, 21 Dec 2015 08:07:57 +0000 (09:07 +0100)]
ntlmssp.idl: MsAvRestrictions is MsvAvSingleHost now

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
8 years agosecurity.idl: add LSAP_TOKEN_INFO_INTEGRITY
Stefan Metzmacher [Mon, 21 Dec 2015 08:06:56 +0000 (09:06 +0100)]
security.idl: add LSAP_TOKEN_INFO_INTEGRITY

This is used in [MS-KILE] and implicit in [MS-NLMP].

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
8 years agoauth/ntlmssp: use ntlmssp_version_blob() in the server
Stefan Metzmacher [Tue, 24 Nov 2015 13:07:23 +0000 (14:07 +0100)]
auth/ntlmssp: use ntlmssp_version_blob() in the server

We already set NTLMSSP_NEGOTIATE_VERSION in
gensec_ntlmssp_server_start(), so it's always
set in chal_flags.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
8 years agoauth/ntlmssp: let the client always include NTLMSSP_NEGOTIATE_VERSION
Stefan Metzmacher [Fri, 20 Nov 2015 09:52:29 +0000 (10:52 +0100)]
auth/ntlmssp: let the client always include NTLMSSP_NEGOTIATE_VERSION

This matches a modern Windows client.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
8 years agoauth/ntlmssp: add ntlmssp_version_blob()
Stefan Metzmacher [Tue, 24 Nov 2015 13:05:17 +0000 (14:05 +0100)]
auth/ntlmssp: add ntlmssp_version_blob()

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
8 years agoauth/ntlmssp: don't send domain and workstation in the NEGOTIATE_MESSAGE
Stefan Metzmacher [Fri, 20 Nov 2015 09:52:29 +0000 (10:52 +0100)]
auth/ntlmssp: don't send domain and workstation in the NEGOTIATE_MESSAGE

We don't set NTLMSSP_NEGOTIATE_OEM_{DOMAIN,WORKSTATION}_SUPPLIED anyway.

This matches modern Windows clients.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
8 years agoauth/ntlmssp: set NTLMSSP_ANONYMOUS for anonymous authentication
Stefan Metzmacher [Tue, 8 Dec 2015 12:59:42 +0000 (13:59 +0100)]
auth/ntlmssp: set NTLMSSP_ANONYMOUS for anonymous authentication

This matches a modern Windows client.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
8 years agoauth/ntlmssp: define all client neg_flags in gensec_ntlmssp_client_start()
Stefan Metzmacher [Tue, 1 Dec 2015 10:01:24 +0000 (11:01 +0100)]
auth/ntlmssp: define all client neg_flags in gensec_ntlmssp_client_start()

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
8 years agoauth/ntlmssp: NTLMSSP_NEGOTIATE_VERSION is not a negotiated option
Stefan Metzmacher [Fri, 20 Nov 2015 09:52:29 +0000 (10:52 +0100)]
auth/ntlmssp: NTLMSSP_NEGOTIATE_VERSION is not a negotiated option

NTLMSSP_NEGOTIATE_VERSION only indicates the existence of the version
information in the packet.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
8 years agoauth/ntlmssp: split out a debug_ntlmssp_flags_raw() that's more complete
Stefan Metzmacher [Tue, 1 Dec 2015 10:16:02 +0000 (11:16 +0100)]
auth/ntlmssp: split out a debug_ntlmssp_flags_raw() that's more complete

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
8 years agos3:ntlm_auth: also use gensec for "ntlmssp-client-1" and "gss-spnego-client"
Stefan Metzmacher [Wed, 9 Dec 2015 09:54:56 +0000 (10:54 +0100)]
s3:ntlm_auth: also use gensec for "ntlmssp-client-1" and "gss-spnego-client"

This implicitly fixes bug #10708.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=10708

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
8 years agowinbindd: make use of ntlmssp_resume_ccache backend for WINBINDD_CCACHE_NTLMAUTH
Stefan Metzmacher [Thu, 26 Nov 2015 10:46:52 +0000 (11:46 +0100)]
winbindd: make use of ntlmssp_resume_ccache backend for WINBINDD_CCACHE_NTLMAUTH

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
8 years agos3:auth_generic: add "ntlmssp_resume_ccache" backend in auth_generic_client_prepare()
Stefan Metzmacher [Thu, 10 Dec 2015 14:42:51 +0000 (15:42 +0100)]
s3:auth_generic: add "ntlmssp_resume_ccache" backend in auth_generic_client_prepare()

This will be used by winbindd in order to correctly implement WINBINDD_CCACHE_NTLMAUTH.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
8 years agoauth/ntlmssp: implement GENSEC_FEATURE_NTLM_CCACHE
Stefan Metzmacher [Fri, 27 Nov 2015 14:35:40 +0000 (15:35 +0100)]
auth/ntlmssp: implement GENSEC_FEATURE_NTLM_CCACHE

This can used in order to use the WINBINDD_CCACHE_NTLMAUTH
code of winbindd to do NTLMSSP authentication with a cached
password.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
8 years agoauth/gensec: add GENSEC_FEATURE_NTLM_CCACHE define
Stefan Metzmacher [Fri, 27 Nov 2015 12:42:30 +0000 (13:42 +0100)]
auth/gensec: add GENSEC_FEATURE_NTLM_CCACHE define

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
8 years agoauth/ntlmssp: provide a "ntlmssp_resume_ccache" backend
Stefan Metzmacher [Wed, 25 Nov 2015 20:41:23 +0000 (21:41 +0100)]
auth/ntlmssp: provide a "ntlmssp_resume_ccache" backend

These can be used to implement the winbindd side of
the WINBINDD_CCACHE_NTLMAUTH call.

It can properly get the initial NEGOTIATE messages
injected if available.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
8 years agos3:ntlmssp: remove unused libsmb/ntlmssp_wrap.c
Stefan Metzmacher [Fri, 11 Dec 2015 11:47:40 +0000 (12:47 +0100)]
s3:ntlmssp: remove unused libsmb/ntlmssp_wrap.c

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
8 years agos3:auth_generic: make use of the top level NTLMSSP client code
Stefan Metzmacher [Thu, 26 Nov 2015 10:45:33 +0000 (11:45 +0100)]
s3:auth_generic: make use of the top level NTLMSSP client code

There's no reason to use gensec_ntlmssp3_client_ops, the
WINBINDD_CCACHE_NTLMAUTH isn't available via gensec anyway.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
8 years agowinbindd: pass an memory context to do_ntlm_auth_with_stored_pw()
Stefan Metzmacher [Tue, 15 Dec 2015 08:07:33 +0000 (09:07 +0100)]
winbindd: pass an memory context to do_ntlm_auth_with_stored_pw()

We should avoid using NULL.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
8 years agos3:tests/test_ntlm_auth_s3: test ntlmssp-client-1 with cached credentials
Stefan Metzmacher [Wed, 9 Dec 2015 15:15:13 +0000 (16:15 +0100)]
s3:tests/test_ntlm_auth_s3: test ntlmssp-client-1 with cached credentials

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11776

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
8 years agos3:torture/test_ntlm_auth.py: add --client-use-cached-creds option
Stefan Metzmacher [Wed, 9 Dec 2015 15:15:13 +0000 (16:15 +0100)]
s3:torture/test_ntlm_auth.py: add --client-use-cached-creds option

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11776

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
8 years agos3:torture/test_ntlm_auth.py: replace tabs with whitespaces
Stefan Metzmacher [Fri, 11 Dec 2015 11:11:05 +0000 (12:11 +0100)]
s3:torture/test_ntlm_auth.py: replace tabs with whitespaces

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
8 years agos3:ntlm_auth: fix --use-cached-creds with ntlmssp-client-1
Stefan Metzmacher [Wed, 9 Dec 2015 20:23:33 +0000 (21:23 +0100)]
s3:ntlm_auth: fix --use-cached-creds with ntlmssp-client-1

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11776

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
8 years agoauth/ntlmssp: add gensec_ntlmssp_server_domain()
Stefan Metzmacher [Wed, 2 Mar 2016 11:06:50 +0000 (12:06 +0100)]
auth/ntlmssp: add gensec_ntlmssp_server_domain()

This is a hack in order to temporary export the server domain
from NTLMSSP through the gensec stack.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agoauth/ntlmssp: keep ntlmssp_state->server.netbios_domain on the correct talloc context
Stefan Metzmacher [Wed, 2 Mar 2016 21:15:50 +0000 (22:15 +0100)]
auth/ntlmssp: keep ntlmssp_state->server.netbios_domain on the correct talloc context

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agos3:auth_generic: add auth_generic_client_start_by_sasl()
Stefan Metzmacher [Tue, 1 Mar 2016 18:39:04 +0000 (19:39 +0100)]
s3:auth_generic: add auth_generic_client_start_by_sasl()

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agos3:auth_generic: add auth_generic_client_start_by_name()
Stefan Metzmacher [Thu, 26 Nov 2015 10:44:02 +0000 (11:44 +0100)]
s3:auth_generic: add auth_generic_client_start_by_name()

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
8 years agoauth/gensec: make gensec_security_by_name() public
Stefan Metzmacher [Thu, 26 Nov 2015 10:43:02 +0000 (11:43 +0100)]
auth/gensec: make gensec_security_by_name() public

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
8 years agoauth/gensec: handle gensec_security_by_sasl_name(NULL, ...)
Stefan Metzmacher [Tue, 1 Mar 2016 18:29:40 +0000 (19:29 +0100)]
auth/gensec: handle gensec_security_by_sasl_name(NULL, ...)

We do that for all other gensec_security_by_*() functions already.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agoauth/gensec: keep a pointer to a possible child/sub gensec_security context
Stefan Metzmacher [Wed, 2 Mar 2016 11:06:50 +0000 (12:06 +0100)]
auth/gensec: keep a pointer to a possible child/sub gensec_security context

This is a hack in order to temporary implement something like:
gensec_ntlmssp_server_domain(), which may be used within spnego.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agos4:pygensec: make sig_size() and sign/check_packet() available
Stefan Metzmacher [Wed, 19 Aug 2015 08:53:34 +0000 (10:53 +0200)]
s4:pygensec: make sig_size() and sign/check_packet() available

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agos3:librpc/gse: implement gensec_gse_max_{input,wrapped}_size()
Stefan Metzmacher [Sat, 5 Mar 2016 01:52:29 +0000 (02:52 +0100)]
s3:librpc/gse: implement gensec_gse_max_{input,wrapped}_size()

This is important in order to support gensec_[un]wrap() with GENSEC_SEAL.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agos3:librpc/gse: don't log gss_acquire_creds failed at level 0
Stefan Metzmacher [Wed, 2 Mar 2016 06:42:41 +0000 (07:42 +0100)]
s3:librpc/gse: don't log gss_acquire_creds failed at level 0

Some callers just retry after a kinit.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agos3:librpc/gse: correctly support GENSEC_FEATURE_SESSION_KEY
Stefan Metzmacher [Tue, 1 Mar 2016 16:37:38 +0000 (17:37 +0100)]
s3:librpc/gse: correctly support GENSEC_FEATURE_SESSION_KEY

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agos3:librpc/gse: set GSS_KRB5_CRED_NO_CI_FLAGS_X in gse_init_client() if available
Stefan Metzmacher [Mon, 22 Jun 2015 13:22:44 +0000 (15:22 +0200)]
s3:librpc/gse: set GSS_KRB5_CRED_NO_CI_FLAGS_X in gse_init_client() if available

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agos3:librpc/gse: fix debug message in gse_init_client()
Stefan Metzmacher [Mon, 22 Jun 2015 13:21:53 +0000 (15:21 +0200)]
s3:librpc/gse: fix debug message in gse_init_client()

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agos3:librpc/gse: make use of GSS_C_EMPTY_BUFFER in gse_init_client
Stefan Metzmacher [Mon, 22 Jun 2015 13:21:05 +0000 (15:21 +0200)]
s3:librpc/gse: make use of GSS_C_EMPTY_BUFFER in gse_init_client

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agowscript_configure_system_mitkrb5: add configure checks for GSS_KRB5_CRED_NO_CI_FLAGS_X
Stefan Metzmacher [Mon, 22 Jun 2015 13:18:22 +0000 (15:18 +0200)]
wscript_configure_system_mitkrb5: add configure checks for GSS_KRB5_CRED_NO_CI_FLAGS_X

Newer MIT versions (maybe krb5-1.14) will also support this.

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agos3:libads: remove unused ads_connect_gc()
Stefan Metzmacher [Wed, 2 Mar 2016 13:36:14 +0000 (14:36 +0100)]
s3:libads: remove unused ads_connect_gc()

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agos4:librpc/rpc: map alter context SEC_PKG_ERROR to NT_STATUS_LOGON_FAILURE
Stefan Metzmacher [Wed, 23 Dec 2015 10:06:47 +0000 (11:06 +0100)]
s4:librpc/rpc: map alter context SEC_PKG_ERROR to NT_STATUS_LOGON_FAILURE

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agolibrpc/rpc: add error mappings for NO_CALL_ACTIVE, OUT_OF_RESOURCES and BAD_STUB_DATA
Stefan Metzmacher [Fri, 17 Jul 2015 01:36:36 +0000 (03:36 +0200)]
librpc/rpc: add error mappings for NO_CALL_ACTIVE, OUT_OF_RESOURCES and BAD_STUB_DATA

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
8 years agodcerpc.idl: make WERROR RPC faults available in ndr_print output
Stefan Metzmacher [Fri, 17 Jul 2015 01:35:19 +0000 (03:35 +0200)]
dcerpc.idl: make WERROR RPC faults available in ndr_print output

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
8 years agoepmapper.idl: make epm_twr_t available in python bindings
Stefan Metzmacher [Thu, 16 Jul 2015 15:15:24 +0000 (17:15 +0200)]
epmapper.idl: make epm_twr_t available in python bindings

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agos3:selftest: run samba3.blackbox.smbclient_auth.plain also with $SERVER_IPV6
Stefan Metzmacher [Tue, 8 Mar 2016 14:53:21 +0000 (15:53 +0100)]
s3:selftest: run samba3.blackbox.smbclient_auth.plain also with $SERVER_IPV6

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agos3:test_smbclient_auth.sh: test using the ip address in the unc path (incl. ipv6...
Stefan Metzmacher [Tue, 8 Mar 2016 14:47:59 +0000 (15:47 +0100)]
s3:test_smbclient_auth.sh: test using the ip address in the unc path (incl. ipv6-literal.net)

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agolib/util_net: add support for .ipv6-literal.net
Stefan Metzmacher [Fri, 4 Mar 2016 01:18:38 +0000 (02:18 +0100)]
lib/util_net: add support for .ipv6-literal.net

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agolib/util_net: move ipv6 linklocal handling into interpret_string_addr_internal()
Stefan Metzmacher [Fri, 4 Mar 2016 01:18:38 +0000 (02:18 +0100)]
lib/util_net: move ipv6 linklocal handling into interpret_string_addr_internal()

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Günther Deschner <gd@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agos4-selftest: Make export keytab test heimdal specific
Andreas Schneider [Tue, 1 Mar 2016 14:54:32 +0000 (15:54 +0100)]
s4-selftest: Make export keytab test heimdal specific

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
8 years agos4-libnet: Implement export_keytab without HDB
Andreas Schneider [Mon, 29 Feb 2016 14:12:02 +0000 (15:12 +0100)]
s4-libnet: Implement export_keytab without HDB

This is used by 'samba-tool domain exportkeytab'. This loads the HDB
Samba backend thus needs access to samdb. To avoid using heimdal
specific code here, we could talk to samdb directly and write a
keytab file.

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
8 years agos3-libnet: Allow the keytab function to use a relative path
Andreas Schneider [Tue, 8 Mar 2016 16:08:22 +0000 (17:08 +0100)]
s3-libnet: Allow the keytab function to use a relative path

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
8 years agokrb5_wrap: Add smb_krb5_open_keytab_relative() function
Andreas Schneider [Tue, 8 Mar 2016 16:07:23 +0000 (17:07 +0100)]
krb5_wrap: Add smb_krb5_open_keytab_relative() function

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
8 years agokrb5_wrap: Move smb_krb5_kt_add_entry() to krb5_wrap
Andreas Schneider [Mon, 29 Feb 2016 16:31:56 +0000 (17:31 +0100)]
krb5_wrap: Move smb_krb5_kt_add_entry() to krb5_wrap

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
8 years agos3-libads: Use the C99 boolean false
Andreas Schneider [Mon, 29 Feb 2016 16:25:33 +0000 (17:25 +0100)]
s3-libads: Use the C99 boolean false

This is a preparation to move smb_krb5_kt_add_entry() to krb5_wrap.

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
8 years agos3-libads: Call smb_krb5_create_key_from_string() directly
Andreas Schneider [Mon, 29 Feb 2016 16:22:50 +0000 (17:22 +0100)]
s3-libads: Call smb_krb5_create_key_from_string() directly

This is a preparation to move smb_krb5_kt_add_entry() to krb5_wrap.

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
8 years agos3-libads: Pass down the salt principal in smb_krb5_kt_add_entry()
Andreas Schneider [Mon, 29 Feb 2016 15:21:56 +0000 (16:21 +0100)]
s3-libads: Pass down the salt principal in smb_krb5_kt_add_entry()

This is a preparation to move smb_krb5_kt_add_entry() to krb5_wrap.

Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
8 years agoCVE-2016-0771: tests/dns: Remove dependencies on env variables
Garming Sam [Fri, 29 Jan 2016 04:28:54 +0000 (17:28 +1300)]
CVE-2016-0771: tests/dns: Remove dependencies on env variables

Now that it is invoked as a normal script, there should be less of them.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11128
BUG: https://bugzilla.samba.org/show_bug.cgi?id=11686

Signed-off-by: Garming Sam <garming@catalyst.net.nz>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agoCVE-2016-0771: tests/dns: change samba.tests.dns from being a unittest
Garming Sam [Fri, 29 Jan 2016 04:03:56 +0000 (17:03 +1300)]
CVE-2016-0771: tests/dns: change samba.tests.dns from being a unittest

This makes it easier to invoke, particularly against Windows.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11128
BUG: https://bugzilla.samba.org/show_bug.cgi?id=11686

Signed-off-by: Garming Sam <garming@catalyst.net.nz>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agoCVE-2016-0771: tests: rename test getopt to get_opt
Garming Sam [Thu, 21 Jan 2016 22:35:03 +0000 (11:35 +1300)]
CVE-2016-0771: tests: rename test getopt to get_opt

This avoids any conflicts in this directory with the original toplevel
getopt.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11128
BUG: https://bugzilla.samba.org/show_bug.cgi?id=11686

Signed-off-by: Garming Sam <garming@catalyst.net.nz>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agoCVE-2016-0771: tests/dns: RPC => DNS roundtrip test
Garming Sam [Wed, 27 Jan 2016 23:54:58 +0000 (12:54 +1300)]
CVE-2016-0771: tests/dns: RPC => DNS roundtrip test

Make sure that TXT entries stored via RPC come out the same in DNS.

This has one caveat in that adding over RPC in Windows eats slashes,
and so fails there.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11128
BUG: https://bugzilla.samba.org/show_bug.cgi?id=11686

Signed-off-by: Garming Sam <garming@catalyst.net.nz>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agoCVE-2016-0771: dnsserver: don't force UTF-8 for TXT
Garming Sam [Wed, 27 Jan 2016 23:36:43 +0000 (12:36 +1300)]
CVE-2016-0771: dnsserver: don't force UTF-8 for TXT

While using a charset is not entirely logical, it allows testing of non
UTF-8 data (like inserting 0xFF into the TXT string).

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11128
BUG: https://bugzilla.samba.org/show_bug.cgi?id=11686

Signed-off-by: Garming Sam <garming@catalyst.net.nz>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agoCVE-2016-0771: tests/dns: modify tests to check via RPC
Garming Sam [Wed, 27 Jan 2016 04:41:44 +0000 (17:41 +1300)]
CVE-2016-0771: tests/dns: modify tests to check via RPC

This checks that TXT records added over DNS, look the same over RPC.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11128
BUG: https://bugzilla.samba.org/show_bug.cgi?id=11686

Signed-off-by: Garming Sam <garming@catalyst.net.nz>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agoCVE-2016-0771: tests/dns: Add some more test cases for TXT records
Garming Sam [Sun, 17 Jan 2016 23:39:46 +0000 (12:39 +1300)]
CVE-2016-0771: tests/dns: Add some more test cases for TXT records

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11128
BUG: https://bugzilla.samba.org/show_bug.cgi?id=11686

Signed-off-by: Garming Sam <garming@catalyst.net.nz>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agoCVE-2016-0771: tests/dns: Correct error code for formerly unrun test
Garming Sam [Wed, 20 Jan 2016 21:25:44 +0000 (10:25 +1300)]
CVE-2016-0771: tests/dns: Correct error code for formerly unrun test

Both Samba and Windows returned NXRRSET

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11128
BUG: https://bugzilla.samba.org/show_bug.cgi?id=11686

Signed-off-by: Garming Sam <garming@catalyst.net.nz>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agoCVE-2016-0771: tests/dns: restore formerly segfaulting test
Garming Sam [Tue, 15 Dec 2015 04:22:32 +0000 (17:22 +1300)]
CVE-2016-0771: tests/dns: restore formerly segfaulting test

This was on the client side, due the a strlen(NULL) on the previously
DOS-encoded TXT field. With a new IDL structure, this segfault no longer exists.
Note that both Samba and Windows return NXRRSET instead of FORMERR.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11128
BUG: https://bugzilla.samba.org/show_bug.cgi?id=11686

Signed-off-by: Garming Sam <garming@catalyst.net.nz>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agoCVE-2016-0771: tests/dns: Add a comment regarding odd Windows behaviour
Garming Sam [Thu, 21 Jan 2016 04:08:18 +0000 (17:08 +1300)]
CVE-2016-0771: tests/dns: Add a comment regarding odd Windows behaviour

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11128
BUG: https://bugzilla.samba.org/show_bug.cgi?id=11686

Signed-off-by: Garming Sam <garming@catalyst.net.nz>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agoCVE-2016-0771: tests/dns: FORMERR can simply timeout against Windows
Garming Sam [Thu, 21 Jan 2016 02:43:55 +0000 (15:43 +1300)]
CVE-2016-0771: tests/dns: FORMERR can simply timeout against Windows

Two requests with identical parameters which are poorly formatted, can
non-deterministically return FORMERR or simply fail to give a response.

Setting the timeout to a number allows Windows to succeed.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11128
BUG: https://bugzilla.samba.org/show_bug.cgi?id=11686

Signed-off-by: Garming Sam <garming@catalyst.net.nz>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agoCVE-2016-0771: tests/dns: prepare script for further testing
Garming Sam [Thu, 21 Jan 2016 03:58:40 +0000 (16:58 +1300)]
CVE-2016-0771: tests/dns: prepare script for further testing

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11128
BUG: https://bugzilla.samba.org/show_bug.cgi?id=11686

Signed-off-by: Garming Sam <garming@catalyst.net.nz>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agoCVE-2016-0771: tests/dns: Modify dns tests to match new IDL
Garming Sam [Wed, 6 Jan 2016 01:12:35 +0000 (14:12 +1300)]
CVE-2016-0771: tests/dns: Modify dns tests to match new IDL

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11128
BUG: https://bugzilla.samba.org/show_bug.cgi?id=11686

Signed-off-by: Garming Sam <garming@catalyst.net.nz>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agoCVE-2016-0771: dns.idl: make use of dnsp_hinfo
Stefan Metzmacher [Fri, 7 Aug 2015 09:36:47 +0000 (11:36 +0200)]
CVE-2016-0771: dns.idl: make use of dnsp_hinfo

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11128
BUG: https://bugzilla.samba.org/show_bug.cgi?id=11686

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agoCVE-2016-0771: s4:dns_server: fix idl for dns_txt_record
Stefan Metzmacher [Fri, 7 Aug 2015 09:36:47 +0000 (11:36 +0200)]
CVE-2016-0771: s4:dns_server: fix idl for dns_txt_record

From RFC 1035:

    3.3.14. TXT RDATA format

        +--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+
        /                   TXT-DATA                    /
        +--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+

    where:

    TXT-DATA        One or more <character-string>s.

    TXT RRs are used to hold descriptive text.  The semantics of the text
    depends on the domain where it is found.

Each record contains an array of strings instead of just one string.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11128
BUG: https://bugzilla.samba.org/show_bug.cgi?id=11686

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agoCVE-2016-0771: librpc: add ndr_dnsp_string_list_copy() helper function
Stefan Metzmacher [Fri, 7 Aug 2015 09:36:47 +0000 (11:36 +0200)]
CVE-2016-0771: librpc: add ndr_dnsp_string_list_copy() helper function

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11128
BUG: https://bugzilla.samba.org/show_bug.cgi?id=11686

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agoCVE-2016-0771: librpc: add RPC_NDR_DNSSERVER to dcerpc-samba library
Stefan Metzmacher [Fri, 7 Aug 2015 09:36:47 +0000 (11:36 +0200)]
CVE-2016-0771: librpc: add RPC_NDR_DNSSERVER to dcerpc-samba library

RPC_NDR_DNSSERVER is the client interface NDR_DNSP contains just
marshalling helpers.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11128
BUG: https://bugzilla.samba.org/show_bug.cgi?id=11686

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agoCVE-2016-0771: s4:librpc: python_dns and python_dcerpc_dnsp doesn't require client...
Stefan Metzmacher [Fri, 7 Aug 2015 09:36:47 +0000 (11:36 +0200)]
CVE-2016-0771: s4:librpc: python_dns and python_dcerpc_dnsp doesn't require client bindings

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11686
BUG: https://bugzilla.samba.org/show_bug.cgi?id=11128

Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
8 years agoCVE-2015-7560: s3: torture3: Add new POSIX-SYMLINK-EA test.
Jeremy Allison [Thu, 7 Jan 2016 22:26:35 +0000 (14:26 -0800)]
CVE-2015-7560: s3: torture3: Add new POSIX-SYMLINK-EA test.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11648

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>
8 years agoCVE-2015-7560: s3: torture3: Add new POSIX-SYMLINK-ACL test.
Jeremy Allison [Thu, 7 Jan 2016 20:58:34 +0000 (12:58 -0800)]
CVE-2015-7560: s3: torture3: Add new POSIX-SYMLINK-ACL test.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11648

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>
8 years agoCVE-2015-7560: s3: libsmb: Add SMB1-only POSIX cli_posix_setacl() functions. Needed...
Jeremy Allison [Thu, 7 Jan 2016 01:02:52 +0000 (17:02 -0800)]
CVE-2015-7560: s3: libsmb: Add SMB1-only POSIX cli_posix_setacl() functions. Needed for tests.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11648

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>
8 years agoCVE-2015-7560: s3: libsmb: Rename cli_posix_getfaclXX() functions to cli_posix_getacl...
Jeremy Allison [Thu, 7 Jan 2016 01:17:24 +0000 (17:17 -0800)]
CVE-2015-7560: s3: libsmb: Rename cli_posix_getfaclXX() functions to cli_posix_getacl() as they operate on pathnames.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11648

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>
8 years agoCVE-2015-7560: s3: smbd: Refuse to set EA's on a symlink.
Jeremy Allison [Tue, 5 Jan 2016 19:33:48 +0000 (11:33 -0800)]
CVE-2015-7560: s3: smbd: Refuse to set EA's on a symlink.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11648

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>
8 years agoCVE-2015-7560: s3: smbd: Silently return no EA's available on a symlink.
Jeremy Allison [Tue, 5 Jan 2016 19:29:38 +0000 (11:29 -0800)]
CVE-2015-7560: s3: smbd: Silently return no EA's available on a symlink.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11648

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>
8 years agoCVE-2015-7560: s3: smbd: Set return values early, allows removal of code duplication.
Jeremy Allison [Tue, 5 Jan 2016 19:05:48 +0000 (11:05 -0800)]
CVE-2015-7560: s3: smbd: Set return values early, allows removal of code duplication.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11648

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>
8 years agoCVE-2015-7560: s3: smbd: Refuse to get a POSIX ACL on a symlink.
Jeremy Allison [Tue, 5 Jan 2016 19:24:36 +0000 (11:24 -0800)]
CVE-2015-7560: s3: smbd: Refuse to get a POSIX ACL on a symlink.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11648

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>
8 years agoCVE-2015-7560: s3: smbd: Refuse to set a POSIX ACL on a symlink.
Jeremy Allison [Tue, 5 Jan 2016 19:22:12 +0000 (11:22 -0800)]
CVE-2015-7560: s3: smbd: Refuse to set a POSIX ACL on a symlink.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11648

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>
8 years agoCVE-2015-7560: s3: smbd: Refuse to set an ACL from a POSIX file handle on a symlink.
Jeremy Allison [Tue, 5 Jan 2016 18:52:50 +0000 (10:52 -0800)]
CVE-2015-7560: s3: smbd: Refuse to set an ACL from a POSIX file handle on a symlink.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11648

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>
8 years agoCVE-2015-7560: s3: smbd: Refuse to get an ACL from a POSIX file handle on a symlink.
Jeremy Allison [Tue, 5 Jan 2016 18:38:28 +0000 (10:38 -0800)]
CVE-2015-7560: s3: smbd: Refuse to get an ACL from a POSIX file handle on a symlink.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11648

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>
8 years agoCVE-2015-7560: s3: smbd: Add refuse_symlink() function that can be used to prevent...
Jeremy Allison [Tue, 5 Jan 2016 19:18:12 +0000 (11:18 -0800)]
CVE-2015-7560: s3: smbd: Add refuse_symlink() function that can be used to prevent operations on a symlink.

BUG: https://bugzilla.samba.org/show_bug.cgi?id=11648

Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Michael Adam <obnox@samba.org>
8 years agoctdb-tunables: Mark tunable DeferredRebalanceOnNodeAdd obsolete
Martin Schwenke [Wed, 9 Mar 2016 01:06:21 +0000 (12:06 +1100)]
ctdb-tunables: Mark tunable DeferredRebalanceOnNodeAdd obsolete

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Autobuild-User(master): Amitay Isaacs <amitay@samba.org>
Autobuild-Date(master): Thu Mar 10 06:51:46 CET 2016 on sn-devel-144

8 years agoctdb-recoverd: Drop use of DeferredRebalanceOnNodeAdd tunable
Martin Schwenke [Tue, 8 Mar 2016 05:36:04 +0000 (16:36 +1100)]
ctdb-recoverd: Drop use of DeferredRebalanceOnNodeAdd tunable

If set, this was used to setup an IP takeover run on a timer after
certain updates to the public IP address configuration (e.g. "ctdb
addip").

However, "ctdb reloadips" completely manages public IP reconfiguration
and avoids the anomalies that DeferredRebalanceOnNodeAdd was
introduced to work around.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
8 years agoctdb-tools: Drop "ctdb rebalancenode"
Martin Schwenke [Tue, 8 Mar 2016 05:29:56 +0000 (16:29 +1100)]
ctdb-tools: Drop "ctdb rebalancenode"

This was a workaround for trying to ensure public IP addresses are
properly rebalanced after running "ctdb addip" on multiple nodes.
"ctdb reloadips" is a better solution.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
8 years agoctdb-tools: Drop "ctdb rebalanceip"
Martin Schwenke [Tue, 8 Mar 2016 05:20:03 +0000 (16:20 +1100)]
ctdb-tools: Drop "ctdb rebalanceip"

This is undocumented and is not needed.  It was a workaround for
trying to ensure public IP addresses are properly rebalanced after
running "ctdb addip" on multiple nodes.  "ctdb reloadips" is a better
solution.

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
8 years agoctdb-doc: Drop outdated NEWS file
Martin Schwenke [Tue, 8 Mar 2016 05:30:41 +0000 (16:30 +1100)]
ctdb-doc: Drop outdated NEWS file

Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Amitay Isaacs <amitay@gmail.com>
8 years agoctdb-doc: Update ctdb man page
Amitay Isaacs [Tue, 8 Mar 2016 04:23:12 +0000 (15:23 +1100)]
ctdb-doc: Update ctdb man page

Update ctdb statistics and ctdb dbstatistics output.

Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>
8 years agoctdb-doc: Update ctdb man page
Amitay Isaacs [Tue, 8 Mar 2016 04:17:24 +0000 (15:17 +1100)]
ctdb-doc: Update ctdb man page

Do not use obsolete tunables in examples.

Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>
8 years agoctdb-tunables: Fix the implementation of LIST_TUNABLES control
Amitay Isaacs [Tue, 8 Mar 2016 04:12:42 +0000 (15:12 +1100)]
ctdb-tunables: Fix the implementation of LIST_TUNABLES control

Do not assume the first tunable is not obsolete.

Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>
8 years agoctdb-recovery-helper: Get tunables first, so control timeout can be set
Amitay Isaacs [Tue, 8 Mar 2016 03:51:59 +0000 (14:51 +1100)]
ctdb-recovery-helper: Get tunables first, so control timeout can be set

During the recovery process, the timeout value for sending all controls
is decided by RecoverTimeout tunable.  So in the recovery process,
first get the tunables, so the control timeout gets set correctly.

Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>
8 years agoctdb-doc: Add documentation for missing tunables
Amitay Isaacs [Tue, 8 Mar 2016 03:37:41 +0000 (14:37 +1100)]
ctdb-doc: Add documentation for missing tunables

Signed-off-by: Amitay Isaacs <amitay@gmail.com>
Reviewed-by: Martin Schwenke <martin@meltin.net>