13 years agor3159: use easy to recognise file handle numbers for new file, old file and directory
Andrew Tridgell [Sun, 24 Oct 2004 13:40:49 +0000 (13:40 +0000)]
r3159: use easy to recognise file handle numbers for new file, old file and directory
in pvfs_open, to make analysing sniffs easy

13 years agor3158: Use the new tower build functions to generate protocol towers, making it
Jelmer Vernooij [Sun, 24 Oct 2004 13:30:50 +0000 (13:30 +0000)]
r3158: Use the new tower build functions to generate protocol towers, making it
easier to deal with more different transports.

13 years agor3157: Add ncacn_http
Jelmer Vernooij [Sun, 24 Oct 2004 13:22:58 +0000 (13:22 +0000)]
r3157: Add ncacn_http

13 years agor3156: Couple of fixes in the conversion functions between binding structs and
Jelmer Vernooij [Sun, 24 Oct 2004 13:00:51 +0000 (13:00 +0000)]
r3156: Couple of fixes in the conversion functions between binding structs and
protocol towers

13 years agor3155: reformat a delete test
Andrew Tridgell [Sun, 24 Oct 2004 12:55:12 +0000 (12:55 +0000)]
r3155: reformat a delete test

13 years agor3154: pvfs now passes all of BASE-DELETE
Andrew Tridgell [Sun, 24 Oct 2004 12:53:13 +0000 (12:53 +0000)]
r3154: pvfs now passes all of BASE-DELETE

13 years agor3153: pvfs now passes the first 9 of the BASE-DELETE tests
Andrew Tridgell [Sun, 24 Oct 2004 12:39:15 +0000 (12:39 +0000)]
r3153: pvfs now passes the first 9 of the BASE-DELETE tests

13 years agor3152: reformatted some of the delete test code
Andrew Tridgell [Sun, 24 Oct 2004 12:38:32 +0000 (12:38 +0000)]
r3152: reformatted some of the delete test code

13 years agor3150: printing __location__ is more useful than a operation number
Andrew Tridgell [Sun, 24 Oct 2004 10:27:38 +0000 (10:27 +0000)]
r3150: printing __location__ is more useful than a operation number

13 years agor3149: separate the delete on close test into torture/basic/delete.c
Andrew Tridgell [Sun, 24 Oct 2004 09:08:52 +0000 (09:08 +0000)]
r3149: separate the delete on close test into torture/basic/delete.c

13 years agor3148: make --failures work for the BASE-DENY1 and BASE-DENY2 tests
Andrew Tridgell [Sun, 24 Oct 2004 09:08:17 +0000 (09:08 +0000)]
r3148: make --failures work for the BASE-DENY1 and BASE-DENY2 tests

13 years agor3147: added basic share modes support for pvfs (or more precisely, ntcreatex
Andrew Tridgell [Sun, 24 Oct 2004 08:31:41 +0000 (08:31 +0000)]
r3147: added basic share modes support for pvfs (or more precisely, ntcreatex
share_access support).  This is enough for us to pass the BASE-DENY2
test, but is a long way from fully correct share modes.

13 years agor3142: fill in all the ntcreatex response fields explicitly, rather than
Andrew Tridgell [Sat, 23 Oct 2004 06:22:57 +0000 (06:22 +0000)]
r3142: fill in all the ntcreatex response fields explicitly, rather than
zeroing. This makes it clearer what bits are not yet implemented (and
is more valgrind friendly)

13 years agor3136: - Allow specifying socket type when adding smbd service
Jelmer Vernooij [Fri, 22 Oct 2004 10:52:57 +0000 (10:52 +0000)]
r3136: - Allow specifying socket type when adding smbd service
- Make sure a epm_tower struct is completely initialized
- Some more minor fixes

13 years agor3135: split the "create new" logic out from the "open existing" logic in
Andrew Tridgell [Fri, 22 Oct 2004 07:01:32 +0000 (07:01 +0000)]
r3135: split the "create new" logic out from the "open existing" logic in
pvfs_open, and handle the various race conditions that are inherent in
cifs on unix, so we do the best we can when the race happens.

the ntcreatex code is really starting to take shape now

13 years agor3134: use struct idr_context * in tid allocation
Andrew Tridgell [Fri, 22 Oct 2004 06:57:31 +0000 (06:57 +0000)]
r3134: use struct idr_context * in tid allocation

13 years agor3133: - more consistent error checking in rename and setfileinfo
Andrew Tridgell [Fri, 22 Oct 2004 06:55:18 +0000 (06:55 +0000)]
r3133: - more consistent error checking in rename and setfileinfo

- add paranoid checking of device/inode change during open to detect race conditions

13 years agor3132: - fixed a type conflict found by talloc_array_p()
Andrew Tridgell [Fri, 22 Oct 2004 06:53:18 +0000 (06:53 +0000)]
r3132: - fixed a type conflict found by talloc_array_p()

- use struct idr_context * in ipc code

13 years agor3131: - make map_nt_error_from_unix() return NT_STATUS_UNSUCCESSFUL if errno is 0
Andrew Tridgell [Fri, 22 Oct 2004 06:49:27 +0000 (06:49 +0000)]
r3131: - make map_nt_error_from_unix() return NT_STATUS_UNSUCCESSFUL if errno is 0

- more consistent checking for system call return values in simple backend

13 years agor3130: - added a LOCAL-IDTREE test suite
Andrew Tridgell [Fri, 22 Oct 2004 06:46:04 +0000 (06:46 +0000)]
r3130: - added a LOCAL-IDTREE test suite

- made idtree return a "struct idr_context *" instead of a void*

- more efficient idr_remove for ids that are not present (patch from Jim Houston)

13 years agor3129: typo
Volker Lendecke [Fri, 22 Oct 2004 04:48:24 +0000 (04:48 +0000)]
r3129: typo

13 years agor3128: Return the correct error code for a secrets/kerberos login, but
Andrew Bartlett [Fri, 22 Oct 2004 03:02:51 +0000 (03:02 +0000)]
r3128: Return the correct error code for a secrets/kerberos login, but
skipping 'bad encryption type'.

Andrew Bartlett

13 years agor3127: added the initial code for the open files database. Doesn't do
Andrew Tridgell [Fri, 22 Oct 2004 01:19:07 +0000 (01:19 +0000)]
r3127: added the initial code for the open files database. Doesn't do
anything yet, but will soon be the core of the shares modes code.

13 years agor3126: in the brlock code I had used a void* for the brl context as I didn't
Andrew Tridgell [Fri, 22 Oct 2004 01:14:49 +0000 (01:14 +0000)]
r3126: in the brlock code I had used a void* for the brl context as I didn't
want to expose the brl context structure outside the brlock.c
code. Instead, I now use "struct brl_context *" and rely on C being
happy to pass around pointers to unknown structures as long as they
are not dereferenced. I will be interested to see how the build farm
likes this.

13 years agor3125: Store object UUID directly instead of using a pointer (struct dcerpc_binding)
Jelmer Vernooij [Thu, 21 Oct 2004 21:57:30 +0000 (21:57 +0000)]
r3125: Store object UUID directly instead of using a pointer (struct dcerpc_binding)
Let test fail if messaging_init() fails instead of generating segfault in the LOCAL-MESSAGING test

13 years agor3124: Add dcerpc_binding_from_tower()
Jelmer Vernooij [Thu, 21 Oct 2004 21:02:51 +0000 (21:02 +0000)]
r3124: Add dcerpc_binding_from_tower()
Add tests for dcerpc_binding_from_tower()

13 years agor3123: Add dcerpc_binding_build_tower()
Jelmer Vernooij [Thu, 21 Oct 2004 19:54:38 +0000 (19:54 +0000)]
r3123: Add dcerpc_binding_build_tower()
Add local test for testing the functions dcerpc_parse_binding(),
dcerpc_binding_string() and dcerpc_binding_build_tower()

13 years agor3118: Eliminate struct dcesrv_ep_description and replace it with
Jelmer Vernooij [Thu, 21 Oct 2004 17:40:55 +0000 (17:40 +0000)]
r3118: Eliminate struct dcesrv_ep_description and replace it with
struct dcerpc_binding.

13 years agor3115: Bugfixes and extra debug in our kerberos verify code.
Andrew Bartlett [Thu, 21 Oct 2004 15:24:50 +0000 (15:24 +0000)]
r3115: Bugfixes and extra debug in our kerberos verify code.

Andrew Bartlett

13 years agor3114: - More work on merging the various structs that describe endpoints
Jelmer Vernooij [Thu, 21 Oct 2004 12:47:02 +0000 (12:47 +0000)]
r3114: - More work on merging the various structs that describe endpoints
- Add protocol sequence to dcerpc transports (will be used later on)
- Add more transports to the list

13 years agor3113: make us able to generate the PAC with the same align as a MS KDC
Stefan Metzmacher [Thu, 21 Oct 2004 12:36:54 +0000 (12:36 +0000)]
r3113: make us able to generate the PAC with the same align as a MS KDC

(we should fix pidl to handle NDR_ALIGN8 correct as a real fix for this problem)


13 years agor3112: Fix two more instances of epm_towers (sorry, metze!)
Jelmer Vernooij [Thu, 21 Oct 2004 11:39:58 +0000 (11:39 +0000)]
r3112: Fix two more instances of epm_towers (sorry, metze!)

13 years agor3111: Add a few more protocol identifiers, rhs for ncalrpc
Jelmer Vernooij [Thu, 21 Oct 2004 10:52:03 +0000 (10:52 +0000)]
r3111: Add a few more protocol identifiers, rhs for ncalrpc

13 years agor3110: Fix the krb5 client and server, so that it doesn't segfault. There
Andrew Bartlett [Thu, 21 Oct 2004 08:52:01 +0000 (08:52 +0000)]
r3110: Fix the krb5 client and server, so that it doesn't segfault.  There
were also gensec bugs that didn't turn up until we hit error paths in
the krb5 code.

Andrew Bartlett

13 years agor3109: Give krbtgt and our machine account a random password in provision.
Andrew Bartlett [Thu, 21 Oct 2004 07:10:59 +0000 (07:10 +0000)]
r3109: Give krbtgt and our machine account a random password in provision.

Andrew Bartlett

13 years agor3108: try to cope with servers that put FILE_ATTRIBUTE_NONINDEXED on files in the...
Andrew Tridgell [Thu, 21 Oct 2004 06:37:15 +0000 (06:37 +0000)]
r3108: try to cope with servers that put FILE_ATTRIBUTE_NONINDEXED on files in the RAW-OPEN test

13 years agor3107: slight tweak to the openx -> ntcreatex mapping routine. This mapping
Andrew Tridgell [Thu, 21 Oct 2004 06:36:14 +0000 (06:36 +0000)]
r3107: slight tweak to the openx -> ntcreatex mapping routine. This mapping
can never be perfect, as openx can do things that ntcreatex can't, but
with this tweak we get close (the BASE-DENY1 test passes completely,
for example)

13 years agor3106: don't call a tree disconnect in the cifs backend, as during a smbd
Andrew Tridgell [Thu, 21 Oct 2004 06:34:17 +0000 (06:34 +0000)]
r3106: don't call a tree disconnect in the cifs backend, as during a smbd
server shutdown we don't want a synchronous operation which may block
to be called, thus delaying the shutdown.

13 years agor3105: using __location__ instead of __LINE__ to give info on error locations is...
Andrew Tridgell [Thu, 21 Oct 2004 03:49:17 +0000 (03:49 +0000)]
r3105: using __location__ instead of __LINE__ to give info on error locations is more useful

13 years agor3104: My Win2k3 server, with current updates, still sends the LM key for level 6.
Andrew Bartlett [Thu, 21 Oct 2004 01:55:46 +0000 (01:55 +0000)]
r3104: My Win2k3 server, with current updates, still sends the LM key for level 6.

I'm not quite sure what's going on here, but adjust expected values
till I can find out how to reproduce this the other way...

Andrew Bartlett

13 years agor3103: use a destructor to ensure that on abnormnal rpc request termination
Andrew Tridgell [Thu, 21 Oct 2004 01:44:46 +0000 (01:44 +0000)]
r3103: use a destructor to ensure that on abnormnal rpc request termination
it is removed from the pending list.

13 years agor3102: typo
Stefan Metzmacher [Thu, 21 Oct 2004 00:49:39 +0000 (00:49 +0000)]
r3102: typo


13 years agor3101: some minor fixes
Stefan Metzmacher [Thu, 21 Oct 2004 00:47:05 +0000 (00:47 +0000)]
r3101: some minor fixes


13 years agor3100: support 'bin/smbclient //w2k3-101/c$ -U \\administrator@w2k3.vmnet1.vm.base'
Stefan Metzmacher [Thu, 21 Oct 2004 00:26:43 +0000 (00:26 +0000)]
r3100: support 'bin/smbclient //w2k3-101/c$ -U \\administrator@w2k3.vmnet1.vm.base'

we need to send an empty string as userdomain in this case.
fix bug #1317 in the client side


13 years agor3099: implment sldb_ModifyDN()
Stefan Metzmacher [Wed, 20 Oct 2004 23:27:09 +0000 (23:27 +0000)]
r3099: implment sldb_ModifyDN()


13 years agor3098: - fix segfault in sldb_Compare()
Stefan Metzmacher [Wed, 20 Oct 2004 23:25:39 +0000 (23:25 +0000)]
r3098: - fix segfault in sldb_Compare()

- be more verbose on the INVALID_DN errstr


13 years agor3097: - an empty string is a valid DN
Stefan Metzmacher [Wed, 20 Oct 2004 23:12:30 +0000 (23:12 +0000)]
r3097: - an empty string is a valid DN

- detect in valid DN's

- some error handling fixes


13 years agor3096: typo
Stefan Metzmacher [Wed, 20 Oct 2004 22:44:08 +0000 (22:44 +0000)]
r3096: typo


13 years agor3095: - fix a free'ing of msg.dn
Stefan Metzmacher [Wed, 20 Oct 2004 20:48:31 +0000 (20:48 +0000)]
r3095: - fix a free'ing of msg.dn

- reenable index tests


13 years agor3094: import all LDAP error codes from the RFC 2251
Stefan Metzmacher [Wed, 20 Oct 2004 20:34:32 +0000 (20:34 +0000)]
r3094: import all LDAP error codes from the RFC 2251


13 years agor3093: - implment ldb_rename() and ldbrename
Stefan Metzmacher [Wed, 20 Oct 2004 19:28:02 +0000 (19:28 +0000)]
r3093: - implment ldb_rename() and ldbrename

- add tests for ldbrename

- disable all tests which regenerate the index
  (this is broken for me...the process hangs,
   tridge we need to discuss that)

- link only the needed stuff to the ldb tools

- build ldbtest inside samba


13 years agor3092: prepare for build inside samba
Stefan Metzmacher [Wed, 20 Oct 2004 19:21:10 +0000 (19:21 +0000)]
r3092: prepare for build inside samba


13 years agor3091: link only the needed stuff
Stefan Metzmacher [Wed, 20 Oct 2004 19:19:03 +0000 (19:19 +0000)]
r3091: link only the needed stuff


13 years agor3089: fix memleak
Stefan Metzmacher [Wed, 20 Oct 2004 17:42:17 +0000 (17:42 +0000)]
r3089: fix memleak


13 years agor3087: fixed a typo
Andrew Tridgell [Wed, 20 Oct 2004 12:24:31 +0000 (12:24 +0000)]
r3087: fixed a typo

13 years agor3086: fixed smbpid handling in the cifs backend
Andrew Tridgell [Wed, 20 Oct 2004 12:08:40 +0000 (12:08 +0000)]
r3086: fixed smbpid handling in the cifs backend

13 years agor3085: make the RAW-WRITE tests more robust to errors in previous parts of the test
Andrew Tridgell [Wed, 20 Oct 2004 11:13:54 +0000 (11:13 +0000)]
r3085: make the RAW-WRITE tests more robust to errors in previous parts of the test

13 years agor3084: mincnt and maxcnt were the wrong way around in readbraw server code
Andrew Tridgell [Wed, 20 Oct 2004 11:11:55 +0000 (11:11 +0000)]
r3084: mincnt and maxcnt were the wrong way around in readbraw server code

13 years agor3083: fixed a couple of generic mapping errors found with RAW-* and cifs:mapgeneric
Andrew Tridgell [Wed, 20 Oct 2004 11:10:51 +0000 (11:10 +0000)]
r3083: fixed a couple of generic mapping errors found with RAW-* and cifs:mapgeneric

13 years agor3082: added a "cifs:mapgeneric" option, which tells the cifs backend to use
Andrew Tridgell [Wed, 20 Oct 2004 11:08:58 +0000 (11:08 +0000)]
r3082: added a "cifs:mapgeneric" option, which tells the cifs backend to use
the ntvfs_generic mapping functions rather than sending the exact
function asked for. This allows the generic mapping functions to be
tested by comparing the behaviour of smbtorture against two cifs
backend shares, one using "cifs:mapgeneric = true" and the other
"cifs:mapgeneric = False"

13 years agor3081: several updates to ntvfs and server side async request handling in
Andrew Tridgell [Wed, 20 Oct 2004 08:28:31 +0000 (08:28 +0000)]
r3081: several updates to ntvfs and server side async request handling in
preparation for the full share modes and ntcreatex code that I am
working on.

highlights include:

 - changed the way a backend determines if it is allowed to process a
   request asynchronously. The previous method of looking at the
   send_fn caused problems when an intermediate ntvfs module disabled
   it, and the caller then wanted to finished processing using this
   function. The new method is a REQ_CONTROL_MAY_ASYNC flag in
   req->control_flags, which is also a bit easier to read

 - fixed 2 bugs in the readbraw server code. One related to trying to
   answer a readbraw with smb signing (which can't work, and crashed
   our signing code), the second related to error handling, which
   attempted to send a normal SMB error packet, when readbraw must
   send a 0 read reply (as it has no header)

 - added several more ntvfs_generic.c generic mapping functions. This
   means that backends no longer need to implement such esoteric
   functions as SMBwriteunlock() if they don't want to. The backend
   can just request the mapping layer turn it into a write followed by
   an unlock. This makes the backends considerably simpler as they
   only need to implement one style of each function for lock, read,
   write, open etc, rather than the full host of functions that SMB
   provides. A backend can still choose to implement them
   individually, of course, and the CIFS backend does that.

 - simplified the generic structures to make them identical to the
   principal call for several common SMB calls (such as
   RAW_WRITE_GENERIC now being an alias for RAW_WRITE_WRITEX).

 - started rewriting the pvfs_open() code in preparation for the full
   ntcreatex semantics.

 - in pvfs_open and ipc_open, initially allocate the open file
   structure as a child of the request, so on error we don't need to
   clean up. Then when we are going to succeed the open steal the
   pointer into the long term backend context. This makes for much
   simpler error handling (and fixes some bugs)

 - use a destructor in the ipc backend to make sure that everthing is
   cleaned up on receive error conditions.

 - switched the ipc backend to using idtree for fnum allocation

 - in the ntvfs_generic mapping routines, use a allocated secondary
   structure not a stack structure to ensure the request pointer
   remains valid even if the backend replies async.

13 years agor3080: Make the Samba4 SAMR server pass the new, nasty torture test (now that
Andrew Bartlett [Wed, 20 Oct 2004 02:26:59 +0000 (02:26 +0000)]
r3080: Make the Samba4 SAMR server pass the new, nasty torture test (now that
SAMR_FIELD_PASSWORD has been split up).

Andrew Bartlett

13 years agor3079: make code more pretty :-)
Andrew Bartlett [Wed, 20 Oct 2004 02:14:28 +0000 (02:14 +0000)]
r3079: make code more pretty :-)

Andrew Bartlett

13 years agor3078: Allow more things to be set as command line options to provision.
Andrew Bartlett [Wed, 20 Oct 2004 02:13:45 +0000 (02:13 +0000)]
r3078: Allow more things to be set as command line options to provision.

Andrew Bartlett

13 years agor3077: Add initial handling of Account Flags in SAMR user info level 21 and 25.
Andrew Bartlett [Wed, 20 Oct 2004 02:12:52 +0000 (02:12 +0000)]
r3077: Add initial handling of Account Flags in SAMR user info level 21 and 25.

Andrew Bartlett

13 years agor3076: Fix memory leak.
Andrew Bartlett [Wed, 20 Oct 2004 02:11:40 +0000 (02:11 +0000)]
r3076: Fix memory leak.

Andrew Bartlett

13 years agor3075: Initialise (and check for intialisation) of the private pointer to
Andrew Bartlett [Wed, 20 Oct 2004 02:10:46 +0000 (02:10 +0000)]
r3075: Initialise (and check for intialisation) of the private pointer to
ensure we don't segfault on the cleanup from an incomplete schannel

Andrew Bartlett

13 years agor3074: Add in a new 'field present' flag samr.idl for the Account Flags
Andrew Bartlett [Wed, 20 Oct 2004 02:08:36 +0000 (02:08 +0000)]
r3074: Add in a new 'field present' flag samr.idl for the Account Flags
field.  Add torture test for setting this feild - including all the
odd cases (not all the flags 'stick', and not others cannot be

Seperate the two 'password change' flags, and test them both in the
torture code.

Check that the password did change after every password set call.

Andrew Bartlett

13 years agor3073: Fix bug in the handling of null-terminated ASCII strings in RPC.
Andrew Bartlett [Wed, 20 Oct 2004 02:02:00 +0000 (02:02 +0000)]
r3073: Fix bug in the handling of null-terminated ASCII strings in RPC.

Because we didn't count the null terminator, we would not move past it
in the packet.

Andrew Bartlett

13 years agor3064: - use UINT8_MAX and UINT16_MAX instead of hex values for idr_get_new() limits
Andrew Tridgell [Tue, 19 Oct 2004 12:06:01 +0000 (12:06 +0000)]
r3064:  - use UINT8_MAX and UINT16_MAX instead of hex values for idr_get_new() limits

 - change idr_get_new() to use > instead of >= in the limit check

13 years agor3063: our default dos charset is CP850, but some systems don't have that, so
Andrew Tridgell [Tue, 19 Oct 2004 10:15:34 +0000 (10:15 +0000)]
r3063: our default dos charset is CP850, but some systems don't have that, so
as a special case, automatically fall back to ASCII if its not found.

13 years agor3062: handle spaces at the start of options in lp_set_cmdline()
Andrew Tridgell [Tue, 19 Oct 2004 10:02:02 +0000 (10:02 +0000)]
r3062: handle spaces at the start of options in lp_set_cmdline()

13 years agor3061: change a debug to help track down a charset problem
Andrew Tridgell [Tue, 19 Oct 2004 09:53:23 +0000 (09:53 +0000)]
r3061: change a debug to help track down a charset problem

13 years agor3060: Replace magic number with a C99 constant.
Tim Potter [Tue, 19 Oct 2004 07:41:19 +0000 (07:41 +0000)]
r3060: Replace magic number with a C99 constant.

13 years agor3059: completely get rid of the MAX_CONNECTIONS limit, as a idle tree
Andrew Tridgell [Tue, 19 Oct 2004 07:18:56 +0000 (07:18 +0000)]
r3059: completely get rid of the MAX_CONNECTIONS limit, as a idle tree
connect is very cheap now.

13 years agor3058: we don't use the bitmap code any more, delete it
Andrew Tridgell [Tue, 19 Oct 2004 07:10:56 +0000 (07:10 +0000)]
r3058: we don't use the bitmap code any more, delete it

13 years agor3057: - moved the idtree.c code into lib/
Andrew Tridgell [Tue, 19 Oct 2004 07:08:35 +0000 (07:08 +0000)]
r3057: - moved the idtree.c code into lib/

- converted the tid handling to use a idtree instead of bitmaps

13 years agor3056: added a id -> pointer data structure (a type of radix tree). This is
Andrew Tridgell [Tue, 19 Oct 2004 06:39:51 +0000 (06:39 +0000)]
r3056: added a id -> pointer data structure (a type of radix tree). This is
an extremely efficient way of mapping from an integer handle (such as
an open file handle) to a pointer (such as the structure containing
the open file information). The code is taken from lib/idr.c in the
2.6 Linux kernel, and is very fast and space efficient. By using
talloc it even has auto cleanup.

This commit converts the handling of open file handles and open
directory search handles to use the idtree routines. In combination
with talloc destructors, this simplifies the structure handling in the
pvfs backend a lot. For example, we no longer need to keep a linked
list of open directory searches at all, and we no longer need to do
linear scans of the list of open files on most operations.

The end result is that the pvfs code is now extremely scalable. You
can have 10s of thousands of open files and open searches and the code
still runs very fast.

I have also added a small optimisation into the file close path, to
avoid looking in the byte range locking database if we know that there
are no locks outstanding.

13 years agor3055: use talloc_zero_p()
Andrew Tridgell [Tue, 19 Oct 2004 06:32:26 +0000 (06:32 +0000)]
r3055: use talloc_zero_p()

13 years agor3054: use talloc_zero_array_p() in a couple of places
Andrew Tridgell [Tue, 19 Oct 2004 06:31:37 +0000 (06:31 +0000)]
r3054: use talloc_zero_array_p() in a couple of places

13 years agor3053: make the maxfid test use subdirectories, so it doesn't create 64k
Andrew Tridgell [Tue, 19 Oct 2004 06:30:52 +0000 (06:30 +0000)]
r3053: make the maxfid test use subdirectories, so it doesn't create 64k
files in one directory (running the test was very slow) and can clean
up after itself easily.

13 years agor3052: added talloc_zero_p() and talloc_zero_array_p() calls, for allocating zeroed...
Andrew Tridgell [Tue, 19 Oct 2004 06:29:41 +0000 (06:29 +0000)]
r3052: added talloc_zero_p() and talloc_zero_array_p() calls, for allocating zeroed memory

13 years agor3047: Always include a \ again before the pipe name we're opening. Without a
Jelmer Vernooij [Mon, 18 Oct 2004 16:25:36 +0000 (16:25 +0000)]
r3047: Always include a \ again before the pipe name we're opening. Without a
backslash works, but is not like Windows does it.

13 years agor3046: \\PIPE\\ is internal (not actually included on the wire)
Jelmer Vernooij [Mon, 18 Oct 2004 16:07:08 +0000 (16:07 +0000)]
r3046: \\PIPE\\ is internal (not actually included on the wire)

13 years agor3045: Allow object-uuid@... binding strings
Jelmer Vernooij [Mon, 18 Oct 2004 15:33:34 +0000 (15:33 +0000)]
r3045: Allow object-uuid@... binding strings

13 years agor3044: resolve the error code for WERR_DS_OBJ_NOT_FOUND to the name
Stefan Metzmacher [Mon, 18 Oct 2004 15:26:16 +0000 (15:26 +0000)]
r3044: resolve the error code for WERR_DS_OBJ_NOT_FOUND to the name


13 years agor3043: Use binding strings for specifying endpoints. The property for
Jelmer Vernooij [Mon, 18 Oct 2004 15:18:05 +0000 (15:18 +0000)]
r3043: Use binding strings for specifying endpoints. The property for
specifying a endpoint is now also 'endpoint' instead of 'endpoints'. The
default endpoint (if none is specified) is still "ncacn_np:[\\pipe\\ifacename]",
where ifacename is the name of the interface.


  endpoint("ncacn_np:[\\pipe\\rpcecho]", "ncacn_ip_tcp:")
interface rpcecho
void dummy();

dcerpc_binding is now converted to ep_description in the server, but I hope to
completely eliminate ep_description later on.

The eventual goal of all these changes is to make it easier to add
 transports as I'm going to add support for
 ncalrpc (local RPC over named pipes) and ncacn_unix_stream (Unix sockets).

13 years agor3041: a start of a README/HOWTO for the samba4 build system.
Stefan Metzmacher [Mon, 18 Oct 2004 14:55:35 +0000 (14:55 +0000)]
r3041: a start of a README/HOWTO for the samba4 build system.
(still incomplete, but should be a good start...)

can someone look for spelling and grammar mistakes...


13 years agor3040: Add sleeps between operations to nbench. Submitted by
Jim McDonough [Mon, 18 Oct 2004 13:58:41 +0000 (13:58 +0000)]
r3040: Add sleeps between operations to nbench.  Submitted by

13 years agor3039: This solves the problem of async handlers in ntvfs backends not being
Andrew Tridgell [Mon, 18 Oct 2004 13:27:22 +0000 (13:27 +0000)]
r3039: This solves the problem of async handlers in ntvfs backends not being
in the right state when called. For example, when we use the unixuid
handler in the chain of handlers, and a backend decides to continue a
call asynchronously then we need to ensure that the continuation
happens with the right security context.

The solution is to add a new ntvfs operation ntvfs_async_setup(),
which calls all the way down through the layers, setting up anything
that is required, and takes a private pointer. The backend wanting to
make a async calls can use ntvfs_async_setup() to ensure that the
modules above it are called when doing async processing.

13 years agor3036: Add function to pull an array of structures. Abstracts away the
Tim Potter [Mon, 18 Oct 2004 12:18:54 +0000 (12:18 +0000)]
r3036: Add function to pull an array of structures.  Abstracts away the
individual routines in ndr_spoolss_buf.c.

13 years agor3035: if the ntvfs layers prior to us have said that we can't perform an
Andrew Tridgell [Mon, 18 Oct 2004 12:06:21 +0000 (12:06 +0000)]
r3035: if the ntvfs layers prior to us have said that we can't perform an
operation asynchronously (such as the nbench module), then ignore lock
timeouts, as they would make no sense

13 years agor3034: - fixed a bug in message dispatch, when the dispatch function called messagin...
Andrew Tridgell [Mon, 18 Oct 2004 11:47:13 +0000 (11:47 +0000)]
r3034:  - fixed a bug in message dispatch, when the dispatch function called messaging_deregister()

 - added a pvfs_lock_close_pending() hook to remove pending locks on file close

 - fixed the private ptr argument to messaging_deregister() in pvfs_wait

 - fixed a bug in continuing lock requests after a lock that is blocking a pending lock is removed

 - removed bogus brl_unlock() call in lock continue

 - corrected error code for LOCKING_ANDX_CHANGE_LOCKTYPE

 - expanded the lock cancel test suite to test lock cancel by unlock and by close

 - added a testsuite for LOCKING_ANDX_CHANGE_LOCKTYPE

13 years agor3033: Use the C struct directly if we can instead of generating a binding
Jelmer Vernooij [Mon, 18 Oct 2004 11:44:07 +0000 (11:44 +0000)]
r3033: Use the C struct directly if we can instead of generating a binding
string and parsing that.

13 years agor3032: Somewhat stricter syntax for binding strings:
Jelmer Vernooij [Mon, 18 Oct 2004 11:43:26 +0000 (11:43 +0000)]
r3032: Somewhat stricter syntax for binding strings:

 [] is now mandatory
 : after the hostname is no longer allowed

examples of allowed binding strings:


Note that the last two lines are not recognized by smbtorture as a binding
 string yet. dcerpc_parse_binding() does accept them though.

13 years agor3031: added support for lock cancelation, which effectively just triggers an early...
Andrew Tridgell [Mon, 18 Oct 2004 09:16:55 +0000 (09:16 +0000)]
r3031: added support for lock cancelation, which effectively just triggers an early lock timeout

added support for more of the bizarre special lock offset semantics of w2k3

13 years agor3030: added testing of lock cancel, and some more special offsets (locks
Andrew Tridgell [Mon, 18 Oct 2004 08:41:44 +0000 (08:41 +0000)]
r3030: added testing of lock cancel, and some more special offsets (locks
with the top bit set are special)

13 years agor3029: implemented byte range lock timeouts.
Andrew Tridgell [Mon, 18 Oct 2004 07:40:17 +0000 (07:40 +0000)]
r3029: implemented byte range lock timeouts.

This adds a pvfs_wait_message() routine which uses the new messaging
system, event timers and talloc destructors to give a nice generic
async event handling system with a easy to use interface. The
extensions to pvfs_lock.c are based on calls to pvfs_wait_message()

We now pass all of our smbtorture locking tests, although while
writing this code I have thought of some additonal tests that should
be added, particularly for lock cancel operations. I'll work on that

This commit also extends the smbtorture lock tests to test the rather
weird 0xEEFFFFFF locking semantics that I have discovered in
win2003. Win2003 treats the 0xEEFFFFFF boundary as special, and will
give different error codes on either side of it. Locks on both sides
are allowed, the only difference is which error code is given when a
lock is denied. Anyone like to hazard a guess as to why? It has
me stumped.

13 years agor3028: use talloc_free() instead of talloc_unlink(), as the
Andrew Tridgell [Mon, 18 Oct 2004 00:04:46 +0000 (00:04 +0000)]
r3028: use talloc_free() instead of talloc_unlink(), as the
event_context_merge() code leaves the events as grandchildren of the
events context, not children, so talloc_unlink() will not work after
the merge

13 years agor3027: got rid of some configure checks we don't need any more
Andrew Tridgell [Sun, 17 Oct 2004 23:59:56 +0000 (23:59 +0000)]
r3027: got rid of some configure checks we don't need any more