| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
all pending receives, when the transport dies. This is because the
async callback most commonly shuts down the connection (it's the only
reasonable thing to do when it's dead), and that frees the whole
context. That means that if we loop more than once, we'll end up using
freed memory.
(This used to be commit 75d537d3a5e3fc5258ce48bfec0c0ce6160978f6)
|
|
* Move dlinklist.h, smb.h to subsystem-specific directories
* Clean up ads.h and move what is left of it to dsdb/
(only place where it's used)
(This used to be commit f7afa1cb77f3cfa7020b57de12e6003db7cfcc42)
|
|
metze
(This used to be commit 49b96ac44a883c020c69df7a12df154dc4faa4d5)
|
|
(This used to be commit 61c6100617589ac6df4f527877241464cacbf8b3)
|
|
try to include just the BASENAME.h files (containing only structs)
(This used to be commit 3dd477ca5147f28a962b8437e2611a8222d706bd)
|
|
metze
(This used to be commit f2e97983f278211c6d70400ce1f43d6a69df0d8a)
|
|
(This used to be commit 70e7449318aa0e9d2639c76730a7d1683b2f4981)
|
|
(This used to be commit 0aca5fd5130d980d07398f3291d294202aefe3c2)
|
|
we get a error from epoll about disabling events for a file descriptor
that is closed
(This used to be commit f32739307464a1f0c835cff886b8c4b960778900)
|
|
hack for the
winbind "bug" :-)
Volker
(This used to be commit fb9a3c7ef376f289288c71bc47d67f548ddb7194)
|
|
problem is that winbind currently relies on being able to receive on a
smb connection from within the same connections receive routine. This
means it relies on a non-serialised connection, so disable the
serialisation until winbind is fixed.
The correct fix will be to get rid of full_request() in dcerpc.c so
that bind requests can be fully async.
(This used to be commit c4115293d83a4a6d103e049c5832d4bcdc0a9dbc)
|
|
enabling of packet serialisation
(This used to be commit 6a47cd65a8b588f9ddd375c57caaba08281e7cbb)
|
|
- call async callback on error
metze
(This used to be commit 43aa5cffd3fd8bf07b236a039f5146e1e44296c6)
|
|
the associated send queue
- fixed negnowait to not watch for the SMBCLI_REQUEST_SEND state
(This used to be commit d19235ede5d352d0b0373d204f4357dddde5946f)
|
|
e.g when you supply an invalid TID or VUID)
- as we don't yet understand how to check the smb_signing of this
replies, we just ignore the whole packet
abartlet,jra,tridge:
can someone try to find out how to create and verify
the signatures for this replies.
what I noticed is that still use the increment by one for the request,
and later requests are still generated fine, only the generating and verifying
of the ntcancel replies make problems
metze
(This used to be commit e6eb0fd2c2f45d6f612d74c6b527c7b17094c907)
|
|
metze
(This used to be commit 356e7d037cf3fc24844b2efa5071917ea03e6163)
|
|
metze
(This used to be commit 16f2d92618a55188d260cadd144281b325cdacda)
|
|
(This used to be commit ba7864b07eebecd4d4eb2ce515412a49964ae179)
|
|
this was cause of the PANIC on the build farm on metze01
as we use req->out.buffer from the old request in the smb_raw_ntcancel()
and send a 0 TID, that causes our server code to crash
(a fix for the server code will follow)
metze
(This used to be commit 97cd824e44b03178706b108c7a78753faf412e8f)
|
|
(This used to be commit d2b568a1114015839ca59c6f32bde4b06ea81ef9)
|
|
(This used to be commit 0bcea45b15b08cb42d7f6fbbb3a25f73b95f362c)
|
|
packet parsing code. This simplifies the logic in the raw client
library a fair bit
(This used to be commit f8d43f1f67876360e1295d85a3c3702d1d60ed7b)
|
|
(This used to be commit b714ab64fd79d5cabc39779774fae7c3861a84da)
|
|
tridge: I think this is correct, comments?
metze
(This used to be commit e06ca726f3df013d869d943338bc6b7a151cdd3f)
|
|
Thanks to lha for giving me a login on a netbsd machine to see this
(This used to be commit 4e66f682e4f1c31bbe9441a13af2c245db31433d)
|
|
disabled. The main change is to turn off spnego, which cannot work at
all without nt status codes (w2k3 gives a ERRHRD:ERRgeneral error when
you try)
I also modified NT_STATUS_EQUAL() to allow for nt->dos code equality,
but only when nt status codes are disabled in smb.conf. That keeps all
the existing torture code working, while still allowing us to
correctly catch the cases where forced dos error codes are needed
The dos->ntstatus mapping table has been removed completely, as it
doesn't really make sense, is impossible to get right, and with the
new dos status handling isn't needed. When matching a nt status code
to a dos status code it makes far more sense to map from the nt code
to the dos code and compare, rather than the reverse, as the nt->dos
mapping is what windows has to do internally, so there really is a
valid mapping table.
(This used to be commit f21274e07b361ef40fdc0fe23e96f1c9c63a091c)
|
|
codes, controlled
with 'nt status support' option.
- make nt_errstr() display nice strings for dos status codes encoded
using NT_STATUS_DOS()
- no longer map between dos and nt status codes in the client library,
instead return using NT_STATUS_DOS()
- fixed the RAW-CONTEXT test to look for
NT_STATUS_DOS(ERRSRV, ERRbaduid) instead of NT_STATUS_INVALID_HANDLE
(This used to be commit ff5549e87ffae9f062394f30d8fd1ae95b614735)
|
|
- add a request destructor, to make it safe to destroy a pending
request with talloc_free()
(This used to be commit 72c6988767249caa585f37fec4c0afbf41557ec2)
|
|
thinks are not needed. Now to see how this fares on the build farm :)
(This used to be commit 80ffcc650c9c86141507edd8338b97814a85f868)
|
|
(This used to be commit b902ea546d2d1327b23f40ddaeeaa8e7e3662454)
|
|
(This used to be commit 7f54c8a339f36aa43c9340be70ab7f0067593ef2)
|
|
gettimeofday() call
and just use timeval_current() when its actually needed
(This used to be commit 236403cc4dc2924ed6a898acae0bb44cc1688dcc)
|
|
make it possible to add optimisations to the events code such as
keeping the next timed event in a sorted list, and using epoll for
file descriptor events.
I also removed the loop events code, as it wasn't being used anywhere,
and changed timed events to always be one-shot (as adding a new timed
event in the event handler is so easy to do if needed)
(This used to be commit d7b4b6de51342a65bf46fce772d313f92f8d73d3)
|
|
remove an event by calling talloc_free().
(This used to be commit 8f19b6886cc58a56d52aecfc83a175197061e533)
|
|
handle the inverted memory hierarchy that a normal session
establishment gave. The inverted hierarchy came from that fact that
you first establish a socket, then a transport, then a session and
finally a tree. That leads to the socket being at the top of the
memory hierarchy and the tree at the bottom, which makes no sense from
the users point of view, as they want to be able to free the tree and
have everything disappear.
The core problem was that the libcli interface didn't distinguish
between establishing a primary context and a secondary context. If you
establish a 2nd session on a transport then you want the transport to
be referenced by the session, whereas if you establish a primary
session then you want the transport to be a child of the session.
To fix this I have added "parent_ctx" and "primary" arguments to the
libcli intialisation functions. This makes using the library much
easier, and gives us a memory hierarchy that makes much more sense.
I was prompted to do this by a bug in the cifs backend, which was
caused by the socket not being properly torn down on a disconnect due
to the inverted memory hierarchy.
(This used to be commit 5e8fd5f70178992e249805c2e1ddafaf6840739b)
|
|
control of the event, so instead build that into the function. If you
pass NULL as mem_ctx then it leaves it as a child of the events
structure.
(This used to be commit 7f981b9ed96f39027cbfd500f41e0c2be64cbb50)
|
|
complexity was that events didn't automatically cleanup
themselves. This was because the events code was written before we had
talloc destructors, so you needed to call event_remove_XX() to clean
the event out of the event lists from every piece of code that used
events. I have now added automatic event destructors, which in turn
allowed me to simplify a lot of the calling code.
The 2nd source of complexity was caused by the ref_count, which was
needed to cope with event handlers destroying events while handling
them, which meant the linked lists became invalid, so the ref_count ws
used to mark events for later destruction.
The new system is much simpler. I now have a ev->destruction_count,
which is incremented in all event destructors. The event dispatch code
checks for changes to this and handles it.
(This used to be commit a3c7417cfeab429ffb22d5546b205818f531a7b4)
|
|
which will eventually try all resolution methods setup in smb.conf
- only resolution backend at the moment is bcast, which does a
parallel broadcast to all configured network interfaces, and takes
the first reply that comes in (this nicely demonstrates how to do
parallel requests using the async APIs)
- converted all the existing code to use the new resolve_name() api
- removed all the old nmb code (yay!)
(This used to be commit 239c310f255e43dd2d1c2433f666c9faaacbdce3)
|
|
recv side in the same event. That's a bad idea, as the first callback
could decide to destroy the socket.
(This used to be commit bf74ea34fc0e3c31e220c8f5a9217c95f3ca1d52)
|
|
should be there too
(This used to be commit 058ae5527e3daeb50eeea9e0ecee858c84e7e17d)
|
|
encapsulates all the different session setup methods, including the
multi-pass spnego code.
I have hooked this into all the places that previously used the
RAW_SESSSETUP_GENERIC method, and have removed the old
RAW_SESSSETUP_GENERIC code from clisession.c and clitree.c. A nice
side effect is that these two modules are now very simple again, back
to being "raw" session setup handling, which was what was originally
intended.
I have also used this to replace the session setup code in the
smb_composite_connect() code, and used that to build a very simple
replacement for smbcli_tree_full_connection().
As a result, smbclient, smbtorture and all our other SMB connection
code now goes via these composite async functions. That should give
them a good workout!
(This used to be commit 080d0518bc7d6fd4bc3ef783e7d4d2e3275d0799)
|
|
reasonable NT_STATUS values
(This used to be commit b193a9cb0c851a4ec55ad9956a815be93eea35e4)
|
|
- added async support to the negprot client code
- removed two unused parameters from smbcli_full_connection() code
- converted smbclient to use smbcli_full_connection() rather than
reinventing everything itself
(This used to be commit 71cbe2873473e039b4511511302cb63f1c50bce8)
|
|
socket connections. This was complicated by a few factors:
- it meant moving the event context from clitransport to clisocket,
so lots of structures changed
- we need to asynchronously handle connection to lists of port
numbers, not just one port number. The code internally tries each
port in the list in turn, without ever blocking
- the man page on how connect() is supposed to work asynchronously
doesn't work in practice (now why doesn't this surprise me?). The
getsockopt() for SOL_ERROR is supposed to retrieve the error, but
in fact the next (unrelated) connect() call on the same socket also
gets an error, though not the right error. To work around this I
need to tear down the whole socket between each attempted port. I
hate posix.
Note that clisocket.c still does a blocking name resolution call in
smbcli_sock_connect_byname(). That will be fixed when we add the async
NBT resolution code.
Also note that I arranged things so that every SMB connection is now
async internally, so using plain smbclient or smbtorture tests all the
async features of this new code.
(This used to be commit 468f8ebbfdbdf37c757fdc4863626aa9946a8870)
|
|
talloc_size() or talloc_array_p() where appropriate.
also fixed a memory leak in pvfs_copy_file() (failed to free a memory
context)
(This used to be commit 89b74b53546e1570b11b3702f40bee58aed8c503)
|
|
error occurs
(This used to be commit 5e13571e6b9f5eb35f710c2c8bd85b5569665613)
|
|
- change smbcli_read/write to take void * for the buffers to match read(2)/write(2)
all this fixes a lot of gcc-4 warnings
metze
(This used to be commit b94f92bc6637f748d6f7049f4f9a30b0b8d18a7a)
|
|
deferred reply is short-circuited immediately when the file is
closed by another user, allowing it to be opened by the waiting user.
- added a sane set of timeval manipulation routines
- converted all the events code and code that uses it to use struct
timeval instead of time_t, which allows for microsecond resolution
instead of 1 second resolution. This was needed for doing the pvfs
deferred open code, and is why the patch is so big.
(This used to be commit 0d51511d408d91eb5f68a35e980e0875299b1831)
|
|
(This used to be commit c6f486574470a311e0d336c026103f131451e21e)
|
|
ioctl.h)
(This used to be commit b97e395c814762024336c1cf4d7c25be8da5813a)
|
