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.
as a human readable string. The format is designed to be able to be
used as the DN for the WINS database as well, while coping with
arbitrary bytes in the name (except nul bytes)
Add #include "system/time.h" back (it was removed in some of these
places because the definitions were provided by <sys/time.h> on tridge's
platform.)
Andrew Bartlett
less likely that anyone will use pstring for new code
- got rid of winbind_client.h from includes.h. This one triggered a
huge change, as winbind_client.h was including system/filesys.h and
defining the old uint32 and uint16 types, as well as its own
pstring and fstring.
- change the iface_n_*() functions to return a "const char *" instead of a "struct ipv4_addr"
I think that in general we should move towards "const char *" for
all IP addresses, as this makes IPv6 much easier, and is also easier
to debug. Andrew, when you get a chance, could you fix some of the
auth code to use strings for IPs ?
- return a NTSTATUS error on bad name queries and node status instead
of using rcode. This makes the calling code simpler.
- added low level name release code in libcli/nbt/
- use a real IP in the register and wins nbt torture tests, as w2k3
WINS server silently rejects some operations that don't come from the
IP being used (eg. it says "yes" to a release, but does not in fact
release the name)
- fixed a bug in the send queue handling on timeouts
- added support for handling unexpected replies (replies to the wrong
port) at the nbtsocket layer
- added separate layer 2 code for wins refresh and wins registration
having the 2nd layer functions do retries themselves. This makes the
code simpler, and allows the TRN_ID to be reused in the retry (which
is how it is supposed to work).
- added support for WACK replies to nbt name requests. A WACK reply
specifies a timeout to wait for the real reply.
- added WINS name refresh async calls, supporting multiple wins
servers and multiple IPs to register
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)
the header, and defined on the wire as a 4 byte network byte order
IP. This means the calling code doesn't have to worry about network
byte order conversions.
event code, as elements of the callers packet structure could go away
while the queue is pending (if for example a name was de-registered
while a packet referencing that name is queued)
broadcast name registration demands per name per interface at 1 second
intervals, then send a name overwrite request and demand. Any name
conflict replies are reported.
names on the network and answers name queries. Lots of details are
still missing, but at least this now means you don't need a Samba3
nmbd to use Samba4.
missing pieces include:
- name registrations should be "shout 3 times, then demand"
- no WINS server yet
- no master browser code
the packets it receives, but it at least shows how the server
structure will work.
To implement it I extended the libcli/nbt/ library to allow for an
incoming packet handler to be registered. That allows the nbt client
library to be used for low level processing of the nbtd server packets.
Other changes:
- made the socket library always set SO_REUSEADDR when binding to an
interface, to ensure that restarts of a server don't have to wait
for a couple of minutes.
- made the nbt port configurable. Defaults to 137, but other ports
will be useful for testing.
asn1-tied-to-blocking-sockets code into the ldap client and torture
suite, and out of the generic libs, so nobody else is tempted to use
it for any new code.
files don't need to match the type names in the generated headers
- with this type mapping we no longer need definitions for the
deprecated "int32", "uint8" etc form of types. We can now force
everyone to use the standard types int32_t, uint8_t etc.
- fixed all the code that used the deprecated types
- converted the IDL types "int64" and "uint64" to "dlong" and
"udlong". These are the 4 byte aligned 64 bit integers that
Microsoft internally define as two 32 bit integers in a
structure. After discussions with Ronnie Sahlberg we decided that
calling these "int64" was confusing, as it implied a true 8 byte
aligned type
- fixed all the cases where we incorrectly used things like
"NTTIME_hyper" in our C code. The generated API now uses a NTTIME for
those. The fact that it is hyper-aligned on the wire is not relevant
to the API, and should remain just a IDL property
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.
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.
