Andrew Tridgell [Mon, 6 Jun 2005 05:52:07 +0000 (05:52 +0000)]
r7322: the beginnings of a in-tree heimdal
Andrew Tridgell [Mon, 6 Jun 2005 05:47:52 +0000 (05:47 +0000)]
r7321: add nbtd statistics serving over irpc
Andrew Tridgell [Mon, 6 Jun 2005 05:47:14 +0000 (05:47 +0000)]
r7320: added support for a private pointer in irpc registered handlers
Derrell Lipman [Mon, 6 Jun 2005 03:01:50 +0000 (03:01 +0000)]
r7316: work in progress
Jelmer Vernooij [Mon, 6 Jun 2005 01:22:57 +0000 (01:22 +0000)]
r7315: Add support for custom makefile targets, e.g.:
[TARGET::FOOBAR]
CMD = make bla
Jelmer Vernooij [Mon, 6 Jun 2005 00:14:23 +0000 (00:14 +0000)]
r7314: the various ncacn and ncadg PDU's share the same namespace
Jelmer Vernooij [Sun, 5 Jun 2005 23:39:00 +0000 (23:39 +0000)]
r7313: Prefix a few functions with ncacn_ rather then dcerpc_ because they are
ncacn_ specific
Jelmer Vernooij [Sun, 5 Jun 2005 23:05:37 +0000 (23:05 +0000)]
r7312: Add IDL for ncadg packets.
Jelmer Vernooij [Sun, 5 Jun 2005 22:59:34 +0000 (22:59 +0000)]
r7311: Use TARGET_CFLAGS for depend list as well
Jelmer Vernooij [Sun, 5 Jun 2005 22:30:20 +0000 (22:30 +0000)]
r7310: Support TARGET_CFLAGS thru recursive make
Andrew Tridgell [Sun, 5 Jun 2005 22:13:27 +0000 (22:13 +0000)]
r7309: started adding IDL for nbt management calls. This adds a
nbtd_information() call that has just one level for now, a
NBTD_STATISTICS level for packet statistics
Jelmer Vernooij [Sun, 5 Jun 2005 16:37:27 +0000 (16:37 +0000)]
r7308: Make dependency lists in the Makefile shorter (see discussion on samba-technical)
Jelmer Vernooij [Sun, 5 Jun 2005 13:52:11 +0000 (13:52 +0000)]
r7307: Add phony rules for OUTPUT_TYPE OBJLIST, e.g. it is now possible
to do things like:
make SUBSYSTEM_NDR_RAW
Andrew Bartlett [Sun, 5 Jun 2005 13:41:32 +0000 (13:41 +0000)]
r7306: Use a consistant #define for detecting support for the Heimdal krb5
log redirection code.
Andrew Bartlett
Jelmer Vernooij [Sun, 5 Jun 2005 13:37:27 +0000 (13:37 +0000)]
r7305: Some cleanups, use write out CFLAGS variable
Andrew Bartlett [Sun, 5 Jun 2005 13:11:42 +0000 (13:11 +0000)]
r7304: Make the libkdc actually work:
- Remove (some) excess logging
- use samdb_connect() to hook into the right handling for multiple tdb handles
- move the connect to the server startup, rather than per-packet.
- Fix config.mk dependency
Tested with a WinXP domain join.
Andrew Bartlett
Stefan Metzmacher [Sun, 5 Jun 2005 12:10:40 +0000 (12:10 +0000)]
r7303: autodetect the libkdc and our kdc support
btw: I use this for configuring heimdal
>>>
CONFIG="CFLAGS=\"-g -O -Wall -Wstrict-prototypes -Wpointer-arith -Wcast-align -Wwrite-strings -Wdeclaration-after-statement\" \
CC=gcc-4.0 \
./configure -C --prefix=$HOME/prefix/heimdal-test \
--sysconfdir=/etc \
--enable-shared=no \
--with-ldb=$HOME/prefix/ldb \
--without-openldap \
--without-openssl $@"
echo $CONFIG
eval $CONFIG
>>>
maybe you also want to use --disable-berkeley-db
metze
Stefan Metzmacher [Sun, 5 Jun 2005 12:06:39 +0000 (12:06 +0000)]
r7302: remove dublicate includes
metze
Stefan Metzmacher [Sun, 5 Jun 2005 08:29:57 +0000 (08:29 +0000)]
r7299: move the list of config.mk files to the source/
so nobody needs to look at build/smb_build/ anymore, just to use the build system
metze
Andrew Tridgell [Sun, 5 Jun 2005 07:44:51 +0000 (07:44 +0000)]
r7298: ensure messages are sent in order even when under extreme load. This
fixes a IO_TIMEOUT problem in the messaging benchmarks
Stefan Metzmacher [Sun, 5 Jun 2005 07:40:17 +0000 (07:40 +0000)]
r7297: make the code more readable
metze
Andrew Tridgell [Sun, 5 Jun 2005 07:37:27 +0000 (07:37 +0000)]
r7296: avoid two stat() calls per message. This increases the raw message
rate from 20k/sec to 55k/sec. The irpc rate goes from 16k/sec to
34k/sec.
I should have run strace -T on this a long time ago.
Andrew Tridgell [Sun, 5 Jun 2005 07:30:44 +0000 (07:30 +0000)]
r7295: added an irpc benchmark. It gets about 16k messages/sec on my laptop,
compared to about 20k messages/sec for the raw messaging layer. I
think that is quite acceptable given the extra functionality.
Andrew Tridgell [Sun, 5 Jun 2005 06:53:07 +0000 (06:53 +0000)]
r7294: implemented the irpc messaging system. This is the core of the
management system I proposed on samba-technical a couple of days
ago. Essentially it is a very lightweight way for any code in Samba to
make IDL based rpc calls to anywhere else in the code, without the
client or server having to go to the trouble of setting up a full rpc
service.
It can be used with any of our existing IDL, but I expect it will
mostly be used for a new set of Samba specific management calls.
The LOCAL-IRPC torture test demonstrates how it can be used by calling
the echo_AddOne() call over this transport.
Andrew Bartlett [Sun, 5 Jun 2005 04:27:12 +0000 (04:27 +0000)]
r7293: Turn sqlite3 support off by default, use --with-sqlite3 to re-enable.
Andrew Bartlett
Andrew Bartlett [Sun, 5 Jun 2005 04:18:57 +0000 (04:18 +0000)]
r7292: Fix up the build system support for derrell's sqlite3 ldb backend.
It is on by default, so I've also fixed a build issue in it. I'll
show how to make it off be default in the next commit.
Andrew Bartlett
Andrew Bartlett [Sun, 5 Jun 2005 04:17:08 +0000 (04:17 +0000)]
r7291: Additional notes on what we require from a kerberos implementation.
Andrew Bartlett
Andrew Tridgell [Sun, 5 Jun 2005 02:20:12 +0000 (02:20 +0000)]
r7290: comment out heimdal config for now
Andrew Tridgell [Sun, 5 Jun 2005 02:00:56 +0000 (02:00 +0000)]
r7289: split out the list of *.mk files for the build, so you don't need to modify the perl
code to add a new library
Jelmer Vernooij [Sun, 5 Jun 2005 00:28:32 +0000 (00:28 +0000)]
r7288: Remove empty wrapper function
Derrell Lipman [Sat, 4 Jun 2005 23:19:41 +0000 (23:19 +0000)]
r7287: work in progress. no more time to work on this right now. save latest changes.
Derrell Lipman [Sat, 4 Jun 2005 22:30:38 +0000 (22:30 +0000)]
r7286: add prototype for public connect function in ldb_sqlite3
Andrew Bartlett [Sat, 4 Jun 2005 22:27:33 +0000 (22:27 +0000)]
r7285: It appears that MIT Kerberos does not have the log redirection
facility that I'm using. This should let us compile the non-KDC
components on MIT again.
Andrew Bartlett
Derrell Lipman [Sat, 4 Jun 2005 21:24:10 +0000 (21:24 +0000)]
r7284: more build issues for sqlite3
Derrell Lipman [Sat, 4 Jun 2005 21:20:56 +0000 (21:20 +0000)]
r7283: add sqlite3.m4
Derrell Lipman [Sat, 4 Jun 2005 21:16:54 +0000 (21:16 +0000)]
r7282: ldb_sqlite3 work in progress.
Derrell Lipman [Sat, 4 Jun 2005 20:30:49 +0000 (20:30 +0000)]
r7281: removing document
Derrell Lipman [Sat, 4 Jun 2005 18:33:21 +0000 (18:33 +0000)]
r7280: taking advantage of previous research. this is documentation on which latest schema is based
Jelmer Vernooij [Sat, 4 Jun 2005 18:05:29 +0000 (18:05 +0000)]
r7278: Oops, don't compile librpc as library..
Derrell Lipman [Sat, 4 Jun 2005 17:13:43 +0000 (17:13 +0000)]
r7276: - moved static tdb function ltdb_dn_fold() into common/ so that it can be
called from multiple backends. (ldb_sqlite3 needs it too.) Added parameter
for a callback function that determines whether an attribute needs case
folding.
- begin to prepare for sqlite3 in build process
- work-in-progress updates, on ldb_sqlite3
Jelmer Vernooij [Sat, 4 Jun 2005 13:57:52 +0000 (13:57 +0000)]
r7274: Move m4 files to seperate dir
Jelmer Vernooij [Sat, 4 Jun 2005 12:29:06 +0000 (12:29 +0000)]
r7273: Update a bit
Andrew Tridgell [Sat, 4 Jun 2005 11:41:20 +0000 (11:41 +0000)]
r7272: this is a sample mk file for building a heimdal library using the
samba4 build system. It assumes we have source/heimdal/ with a current
lorikeet heimdal tree.
I haven't enabled this, its just an example of one possible approach
to heimdal integration.
Andrew Tridgell [Sat, 4 Jun 2005 11:39:23 +0000 (11:39 +0000)]
r7271: added the ability to specify a target specific set of CFLAGS for
binaries and libraries. This makes it possible to build heimdal with
our build system, which means users don't have to suffer two build
systems.
Andrew Bartlett [Sat, 4 Jun 2005 11:17:05 +0000 (11:17 +0000)]
r7270: A big revamp to the way we handle kerberos errors in Samba4. We now
fill in the function pointers to handle the logging, and catch all the
kerberos warnings. (Currently at level 3).
To avoid a memory leak, this requries a new function: krb5_freelog(),
which I've added to lorikeet/heimdal.
This also required a revamp to how we handle the krb5_context, so as
to make it easier to handle with talloc destructors.
Andrew Bartlett
Andrew Bartlett [Sat, 4 Jun 2005 11:10:57 +0000 (11:10 +0000)]
r7269: talloc_steal() is preferred where possible, as it can't fail and does
not have some of the issues of talloc_reference().
Andrew Bartlett
Simo Sorce [Sat, 4 Jun 2005 08:54:07 +0000 (08:54 +0000)]
r7268: allocate the strings to avoid them disappearing under our feet
Simo Sorce [Sat, 4 Jun 2005 08:23:15 +0000 (08:23 +0000)]
r7267: REMOTE_HOST is a better choice
Tim Potter [Sat, 4 Jun 2005 07:04:43 +0000 (07:04 +0000)]
r7266: Split the different types of js function defines into separate files,
as there are going to be a lot more of them.
Andrew Tridgell [Sat, 4 Jun 2005 06:45:11 +0000 (06:45 +0000)]
r7265: fixed d_printf() so it works again.
Jelmer, can you give me a hint as to why you changed this? It
completely broke smbclient, and the change now makes d_printf() almost
equivalent to printf() as you removed the language translation
code. See your commit 7166.
Did you intend to reenable language handling by some other method? Or
do you just mean to remove it completely?
Andrew Tridgell [Sat, 4 Jun 2005 05:35:27 +0000 (05:35 +0000)]
r7264: fix up the socket handling for abartlet. Still only udp, but it won't
be hard to do tcp as well.
Tim Potter [Sat, 4 Jun 2005 03:51:38 +0000 (03:51 +0000)]
r7263: Exit smbscript with the intepreter return value (defaults to 0).
Change the exit value for an exception, usage error and other non-js
errors to 127 which is kinda like the return value for the system(3)
function.
Tim Potter [Sat, 4 Jun 2005 03:35:38 +0000 (03:35 +0000)]
r7262: Add a length property to ARGV array.
Tim Potter [Sat, 4 Jun 2005 03:32:18 +0000 (03:32 +0000)]
r7261: Pass by reference is done in js via MPR_TYPE_OBJECT. Update argument
parsing and example for resolveName().
Derrell Lipman [Sat, 4 Jun 2005 02:31:45 +0000 (02:31 +0000)]
r7260: save current schema before I blow it away to try something different
Andrew Bartlett [Sat, 4 Jun 2005 01:40:30 +0000 (01:40 +0000)]
r7259: Move the recv handler out into a seperate function (suggestion from
tridge)
Andrew Bartlett
Andrew Bartlett [Sat, 4 Jun 2005 01:19:05 +0000 (01:19 +0000)]
r7258: Fix the final linking error with libkdc - we need to link libhdb as well.
With this fix, I can request tickets from our built-in KDC!
Andrew Bartlett
Andrew Bartlett [Sat, 4 Jun 2005 01:17:54 +0000 (01:17 +0000)]
r7257: Ensure the error message can never be uninitialised.
Andrew Bartlett
Tim Potter [Sat, 4 Jun 2005 01:10:21 +0000 (01:10 +0000)]
r7256: Demonstrate use of resolveName() js function.
Tim Potter [Sat, 4 Jun 2005 01:08:52 +0000 (01:08 +0000)]
r7255: Change syntax of resolveName() js function to be more like the
resolve_name() C function. I can't figure out how to return variables
by reference though. Writing to argv[] doesn't seem to work.
Tim Potter [Sat, 4 Jun 2005 01:06:30 +0000 (01:06 +0000)]
r7254: Add a mprWERROR() function with the same attributes as mprNTSTATUS.
Tim Potter [Sat, 4 Jun 2005 00:20:32 +0000 (00:20 +0000)]
r7253: Fix build.
Rafal Szczesniak [Fri, 3 Jun 2005 23:11:51 +0000 (23:11 +0000)]
r7252: Formatting.
rafal
Rafal Szczesniak [Fri, 3 Jun 2005 23:11:09 +0000 (23:11 +0000)]
r7251: Initial work on composite domain open call.
rafal
Rafal Szczesniak [Fri, 3 Jun 2005 23:10:26 +0000 (23:10 +0000)]
r7250: Starting work on composite domain open call.
rafal
Jelmer Vernooij [Fri, 3 Jun 2005 21:30:07 +0000 (21:30 +0000)]
r7249: Cope with struct member rename
Jelmer Vernooij [Fri, 3 Jun 2005 20:44:46 +0000 (20:44 +0000)]
r7248: Remove enum that is causing trouble on AIX
Rafal Szczesniak [Fri, 3 Jun 2005 19:09:21 +0000 (19:09 +0000)]
r7247: User shorter and more convenient name for credentials in
libnet context.
rafal
Simo Sorce [Fri, 3 Jun 2005 14:36:07 +0000 (14:36 +0000)]
r7242: typo
Andrew Bartlett [Fri, 3 Jun 2005 14:32:10 +0000 (14:32 +0000)]
r7241: The KDC almost links...
Using current lorikeet/heimdal, and with the KDC module enabled (it is
disabled by default), I almost get the KDC to link.
(To enable the KDC for testing, comment out the only line in
smbd/config.m4, and add 'kdc' to the 'server services' line in
smb.conf).
Andrew Bartlett [Fri, 3 Jun 2005 14:30:03 +0000 (14:30 +0000)]
r7240: Don't call our fancy error message routines on a null context.
Andrew Bartlett
Simo Sorce [Fri, 3 Jun 2005 14:20:25 +0000 (14:20 +0000)]
r7239: add a little test
Simo Sorce [Fri, 3 Jun 2005 14:17:18 +0000 (14:17 +0000)]
r7238: Add pam auth support in swat
Simo Sorce [Fri, 3 Jun 2005 14:15:06 +0000 (14:15 +0000)]
r7237: we need to check for PAM here
Simo Sorce [Fri, 3 Jun 2005 14:11:04 +0000 (14:11 +0000)]
r7236: uhmm this require to be in too
Simo Sorce [Fri, 3 Jun 2005 14:09:08 +0000 (14:09 +0000)]
r7235: We do not need PAM here anymore
Simo Sorce [Fri, 3 Jun 2005 14:08:27 +0000 (14:08 +0000)]
r7234: that code path is died
Simo Sorce [Fri, 3 Jun 2005 14:07:50 +0000 (14:07 +0000)]
r7233: what about filling that new element ?
Simo Sorce [Fri, 3 Jun 2005 14:06:23 +0000 (14:06 +0000)]
r7232: add some more auth stuff
enables us to authenticate against system users (only PAM support right now)
Simo Sorce [Fri, 3 Jun 2005 14:05:05 +0000 (14:05 +0000)]
r7231: remove some old stuff
Andrew Tridgell [Fri, 3 Jun 2005 13:37:23 +0000 (13:37 +0000)]
r7230: use socket_pending() to get rid of the max packet size limits in the
cldap code
Andrew Tridgell [Fri, 3 Jun 2005 13:31:27 +0000 (13:31 +0000)]
r7229: use socket_pending() to get rid of the max packet size limits in the
nbt and dgram layers
Andrew Tridgell [Fri, 3 Jun 2005 13:20:45 +0000 (13:20 +0000)]
r7228: use socket_pending() instead of the direct ioctl in the messaging code
Andrew Tridgell [Fri, 3 Jun 2005 13:20:08 +0000 (13:20 +0000)]
r7227: added a socket_pending() call to abstract away the FIONREAD ioctl. It
will be interesting to see if this causes any portability problems, as
it is a less commonly used call.
Andrew Bartlett [Fri, 3 Jun 2005 13:04:44 +0000 (13:04 +0000)]
r7226: Forgot file to disable building the new kdc
Andrew Bartlett
Tim Potter [Fri, 3 Jun 2005 12:31:56 +0000 (12:31 +0000)]
r7225: Create a MprVar object from a NTSTATUS, e.g:
res: {
is_err: true,
is_ok: false,
errstr: "NT_STATUS_IO_TIMEOUT",
v: -
1073741643
}
Simo Sorce [Fri, 3 Jun 2005 12:13:33 +0000 (12:13 +0000)]
r7224: add some more usefull data to the auth_usersupplied_info struct
Tim Potter [Fri, 3 Jun 2005 12:04:26 +0000 (12:04 +0000)]
r7223: Advance script past interpreter line.
Simo Sorce [Fri, 3 Jun 2005 11:25:55 +0000 (11:25 +0000)]
r7222: remove double call to the same function in the same place, probable cut&paste error
Andrew Bartlett [Fri, 3 Jun 2005 11:23:15 +0000 (11:23 +0000)]
r7221: Add the start of a KDC service (to be built on a 'libkdc' from a to be
included Heimdal) to Samba4.
Andrew Bartlett
Andrew Bartlett [Fri, 3 Jun 2005 11:19:32 +0000 (11:19 +0000)]
r7220: Fix comment
Andrew Bartlett [Fri, 3 Jun 2005 11:18:48 +0000 (11:18 +0000)]
r7219: Don't allow 'binding' to be used uninitilaised.
Andrew Bartlett
Andrew Bartlett [Fri, 3 Jun 2005 11:18:07 +0000 (11:18 +0000)]
r7218: Don't use an uninitialised variable in an error message.
Andrew Bartlett
Tim Potter [Fri, 3 Jun 2005 08:00:42 +0000 (08:00 +0000)]
r7215: Convert smbscript to use ejsEvalScript() and file_load() instead of
ejsEvalFile().
Still need to add advancement of the script past the hash-bang line but
it's home time now!!
Tim Potter [Fri, 3 Jun 2005 07:56:41 +0000 (07:56 +0000)]
r7214: Undo my interpreter patch. Tridge thought of a much better way to do it
that doesn't involve any changes to ejs at all.
Tim Potter [Fri, 3 Jun 2005 07:47:06 +0000 (07:47 +0000)]
r7213: Add a small bit to the ejs parser to ignore a #!/path/to/interpreter
as the first line of the script. This allows smbscript to be used as an
interpreter.
Tim Potter [Fri, 3 Jun 2005 07:03:26 +0000 (07:03 +0000)]
r7212: Disambiguate dummy functions to avoid linker problemos.
Andrew Tridgell [Fri, 3 Jun 2005 06:04:34 +0000 (06:04 +0000)]
r7211: - use ioctl(FIONREAD) to remove the artificial limit on messaging size
- avoid a memcpy
Andrew Tridgell [Fri, 3 Jun 2005 04:21:25 +0000 (04:21 +0000)]
r7206: changed the messaging library to use sendto instead of a connected
send on the unix domain datagram socket. This gains us about 50% in
speed, and also means that we don't run the risk of running out of
file descriptors due to heavy messaging traffic. We now use a single
file descriptor no matter how many messages are pending to any number
of servers.
Andrew Tridgell [Fri, 3 Jun 2005 04:19:32 +0000 (04:19 +0000)]
r7205: added support for sendto() on unix domain sockets