IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an
email to Administrator. User accounts are meant only to access repo
and report issues and/or generate pull requests.
This is a purpose-specific Git hosting for
BaseALT
projects. Thank you for your understanding!
Только зарегистрированные пользователи имеют доступ к сервису!
Для получения аккаунта, обратитесь к администратору.
This sometimes would show up as corrupted bytes during logs. Hammering
the LDAP server enough times managed to trigger an outright segfault.
Signed-off-by: Garming Sam <garming@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Use the passed in context from callers. Remove one
talloc_autofree_context().
Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
This makes sure we remove any messaging sockets if a child dies or calls exit()
without running the talloc destructor for messaging
Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Not currently used - no logic changes inside.
This will make it possible to pass down a long-lived talloc
context from the loading function for modules to use instead
of having them internally all use talloc_autofree_context()
which is a hidden global.
Updated all known module interface numbers, and added a
WHATSNEW.
Signed-off-by: Jeremy Allison <jra@samba.org>
Signed-off-by: Ralph Böhme <slow@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Sat Apr 22 01:17:00 CEST 2017 on sn-devel-144
Ensure it's freed on all error paths.
Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Mon Apr 17 23:10:06 CEST 2017 on sn-devel-144
Use state->event_ctx as the parent of the initial imessaging context.
Now we control all exit paths, we can call TALLOC_FREE(state)
on all of them.
Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Defensive programming change. Not strictly needed to prevent
any crash/error.
Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Simplify by removing global state we don't need now
we're called by tevent (and in the short window where
we're installed by CatchSignal but before we install
the tevent handler we don't need the complex global
state handling as we have no forked children).
We now have access to struct server_state on all
exit paths - next commits will stop using talloc autofree context.
Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Use it in the message print to avoid a "unused variable" compile error.
Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
No logic changes, will be used to move allocated
pointers off the talloc autofree context in a later commit.
Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Allows us to be independent of parent SIGTERM signal handling.
Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
Autobuild-User(master): Ralph Böhme <slow@samba.org>
Autobuild-Date(master): Sat Apr 8 16:21:57 CEST 2017 on sn-devel-144
In the NTVFS server we have the following talloc heirarchy:
event_ctx
|
---------------------------------------------------- .. other children
| | |
msg_dgm_ref srv_conn msg_dgm_ref
^ |
| NTVFS structures
| |
| XXXXXX
| |
| |
--------------------- pointer to msg_dgm_ref
Some of the structures under NTVFS （marked XXXXX) can have
pointers to imessaging contexts which internally have pointers
to msg_dgm_ref structurs allocated off event_ctx.
The original code calls:
model_ops->terminate(event_ctx, srv_conn->lp_ctx, reason);
talloc_free(srv_conn);
But model_ops->terminate() calls talloc_free(event_ctx) and
then calls exit(). In this case srv_conn is never explicitly
freed, but only freed as a talloc child of the event_ctx.
Depending on the ordering of the linked list of talloc children
under event_ctx（which can be reordered via talloc_free/reinit
of msg_dgm_ref) a pointer to msg_dgm_ref under srv_conn can
be left pointing to memory that was already freed. This pointer
is then used in the destructor for a file object called when
srv_conn is freed.
Re-ordering this to explicitly call TALLOC_FREE(srv_conn) first
and then model_ops->terminate() fixes this problem.
Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Böhme <slow@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Sun Apr 2 05:18:39 CEST 2017 on sn-devel-144
Ensure it is called from process_standard.c after
every fork().
Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Autobuild-User(master): Volker Lendecke <vl@samba.org>
Autobuild-Date(master): Fri Mar 31 14:48:17 CEST 2017 on sn-devel-144
This brings the callers of named_pipe_auth in line with that subsystem.
While these names may be better, the rest of Samba consistently uses
remote_address and local_address, and this difference has hidden bugs
Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Pair-Programmed-by: Gary Lockyer <gary@catalyst.net.nz>
Signed-off-by: Gary Lockyer <gary@catalyst.net.nz>
We need to know in the RPC server the original address the client came from
so that we can log this with the authentication audit information
Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Pair-Programmed-by: Gary Lockyer <gary@catalyst.net.nz>
Signed-off-by: Gary Lockyer <gary@catalyst.net.nz>
Signed-off-by: Chris Lamb <chris@chris-lamb.co.uk>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>
BUG: https://bugzilla.samba.org/show_bug.cgi?id=12521
Pair-Programmed-With: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Andreas Schneider <asn@samba.org>
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Since Windows Server 2008 Microsoft uses a different port range for RPC
services. Before it was 1024-65535 and they changed it to 49152-65535.
We should use the same range as these are the ports the firewall in AD
networks normally allow.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=12521
Signed-off-by: Andreas Schneider <asn@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
This option is unused and has not been used since before Samba 4.3
when the source4/ winbindd code went away.
The associated dynconfig parameters used for the default are also removed.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=10066
Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Autobuild-User(master): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(master): Sun Jan 1 13:47:26 CET 2017 on sn-devel-144
The warning is:
../source4/smbd/service_stream.c: In function ‘stream_io_handler’:
../source4/smbd/service_stream.c:94:18: warning: increment of a boolean expression [-Wbool-operation]
conn->processing++;
^~
../source4/smbd/service_stream.c💯18: warning: decrement of a boolean expression [-Wbool-operation]
conn->processing--;
^~
while the code in question looks like:
conn->processing++;
if (flags & TEVENT_FD_WRITE) {
conn->ops->send_handler(conn, flags);
} else if (flags & TEVENT_FD_READ) {
conn->ops->recv_handler(conn, flags);
}
conn->processing--;
If this is never going to be nested, processing can be bool and the ++
and -- can be true/false assignments. But it seems possible that these
might be nested so it is better to go the other way.
Signed-off-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Reviewed-by: Ralph Boehme <slow@samba.org>
Autobuild-User(master): Ralph Böhme <slow@samba.org>
Autobuild-Date(master): Thu Dec 1 00:28:05 CET 2016 on sn-devel-144
Ensure the messaging dgm context goes away *before* the tevent
context. The messaging dgm context will likely have active fd or timer
events, their rundown will touch the associated tevent context.
Otoh, I deliberately don't free the imessaging context here, that's going
to happen as part of freeing the talloc_autofree_context() as before. I
think it suffers the same problem, eg imessaging_deregister() works on
an imessaging_context that might already be freed. But as it works,
don't change it.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
With modern messaging this doesn't do anything (it's an
empty destructor). Clean up so we can add a proper destructor
in future.
Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
These were never finished, were not tested and clearly will not be revived
Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
We would build, but not use, many components of the NTVFS file server
even when we asked not to. They would then consume disk, but not be
of any use
Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>
This helps us understand failure modes in selftest
Reviewed-by: Douglas Bagnall <douglas.bagnall@catalyst.net.nz>
Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Autobuild-User(master): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(master): Fri Jan 1 03:23:14 CET 2016 on sn-devel-144
This removes quite a bit of code. All reasonable systems have /dev/urandom
these days. Linux, Solaris and the BSDs do. In case we find a system
without /dev/urandom, we will have to go hunting in other libraries.
The main reason for this is speed: On Ubuntu 14.04 doing direct reads from
/dev/urandom is 2-3 times faster than our md4 based code. On virtualized
FreeBSD 10 the difference is even larger.
My first approach was to use fopen/fread. It was even faster, but less
than twice as fast. So I thought we could save the additional complexity
when having to deal with throwing away buffers when forking and the
additional memory footprint per process.
With this simple generate_random_buffer it will be easier to adapt new
syscalls to get randomness.
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Tue Oct 13 04:25:39 CEST 2015 on sn-devel-104
Pair-Programmed-With: Michael Adam <obnox@samba.org>
Signed-off-by: Andreas Schneider <asn@samba.org>
Signed-off-by: Michael Adam <obnox@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
BUG: https://bugzilla.samba.org/show_bug.cgi?id=11180
The issue has been found with nss_wrapper debug output running:
samba4.ntvfs.cifs.krb5.base.lock
In the case here, we fork a child and close the fd without resetting
the pipe fd variable. Then the fd was used to open the nss_wrapper
hosts file which got the same fd. We forked again in the process model
called close() on the re-used fd (of the pipe variable) again without
nss_wrapper noticing. Now Samba opened the secrets tdb and got
the same fd as nss_wrapper was using for the hosts file and next
nss_wrapper tried to parse a TDB ...
Pair-Programmed-With: Michael Adam <obnox@samba.org>
Signed-off-by: Andreas Schneider <asn@samba.org>
Signed-off-by: Michael Adam <obnox@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
We replace this with a pipe between parent and child, and then watch
for a read event in the parent to indicate that the child has gone away.
The removal of signal(SIGCHLD, SIG_IGN) requires us to then call
waitpid(). We can't do that in a main loop as we want to get the exit
status to the legitimate waitpid calls in routines like
samba_runcmd_*().
Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
While it is possible to run Samba in these modes, it isn't tested and
isn't required.
Each of these modes requires that SIGCHLD be set to SIGIGN, and that
breaks samba_runcmd_*().
Signed-off-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Jelmer Vernooij <jelmer@samba.org>
With this you can watch "samba"'s talloc hierarchy live using
smbcontrol <pid> pool-usage
Enjoy :-)
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Autobuild-User(master): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(master): Sat Feb 14 01:59:19 CET 2015 on sn-devel-104
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Autobuild-User(master): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(master): Thu Jan 1 02:47:59 CET 2015 on sn-devel-104
This code has not seen any use or development for a while. While probably
being a good source of inspiration, it will inevitably bit-rot. To really
get to a multi-threaded process model, many of the Samba libraries have
to be made thread-safe.
Why this patch? Right now I am trying to clean up our debug system,
and process_thread.c is the only reference to the _suspicious_ functions
in debug.c. The alternative would be to factor those out of debug, but
as this code is unused for so long, the alternative approach is to just
remove it.
If someone wants to restart development on a threaded process model,
the code is still around in the git history and can be resurrected easily.
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
