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.
Use winbindd to get the group list if possible since we already
know it from netsamlogon_cache.tdb. More effecient than letting
libc call getgrent() to get seconary groups.
Tested by Ken Cross.
(This used to be commit 3c537c906f)
1) don't ask trusted DC's for a list of trusted domains. This causes
us to treat non-transitive ones as if they were transitive. Not
needed anyways
2) Fix dc lookup bug where we would always try to use DNS to resolve
the DC's for a domain (even if it was a trusted NT4 domain).
(This used to be commit 4d3acce506)
I was storing the mid of the oplock break - I should have been
storing the mid from the open. There are thus 2 types of deferred
packet sequence returns - ones that increment the sequence number
(returns from oplock causing opens) and ones that don't (change notify
returns etc). Running with signing forced on does lead to some
interesting tests :-).
Jeremy.
(This used to be commit 85907f02ce)
- remove 'if(mem_ctx)' tests prior to 'talloc_destroy' call to make
consistent with other modules; 'talloc_destroy' already test for NULL
anyway.
- initialize PyObject* result pointers to NULL in function
declarations; enables removal of redundant NULL assignments.
- use local scope TALLOC_CTX in lsa_lookup_names to prevent unbounded
memory growth during python policy object lifetime.
- change context name string used in lsa_lookup_sids from
'lsa_open_policy' to 'lsa_lookup_sids' (cut'npaste oversight from
previous patch)
- change docstring to match module name (apparently another cut'npaste
situation)
(This used to be commit 98cac74c67)
Otherwise we find spurious mid sign records on reply_ntcancel calls (they cancel
by mid). That took a *lot* of tracking down. I still need to remove the mid
records from the sign state on reply_ntcancel to avoid leaking memory....
Jeremy.
(This used to be commit 270bf20fe3)
are updated correctly on returning an error for server trans streams.
Ensure we turn off client trans streams on error.
Jeremy.
(This used to be commit 3a789cb7f0)
bug with w2k. Turns out that when we're doing a trans/trans2/nttrans call
the MID and send_sequence_number and reply_sequence_number must remain constant.
This was something we got very wrong in earlier versions of Samba. I can now
get a directory listing from WINNT\SYSTEM32 with the older earlier parameters
for clilist.c
This still needs to be fixed for the server side of Samba, client appears to
be working happily now (I'm doing a signed smbtar download of an entire W2K3
image to test this :-).
Jeremy.
(This used to be commit 2093a3130d)
queue if the posix lock failed with EACCES or EAGAIN (this means another
lock conflicts). Else return an error and don't queue the request.
Jeremy.
(This used to be commit 43fbc18fdc)
from RFC but I'm smelling a client bug here.
/* only look at the first OID for determining the mechToken --
accoirding to RFC2478, we should choose the one we want
and renegotiate, but i smell a client bug here..
Problem observed when connecting to a member (samba box)
of an AD domain as a user in a Samba domain. Samba member
server sent back krb5/mskrb5/ntlmssp as mechtypes, but the
client (2ksp3) replied with ntlmssp/mskrb5/krb5 and an
NTLMSSP mechtoken. --jerry */
(This used to be commit 731420b03d)
Anybody familiar with Makefile.in could you please look at this?
This is probably the wrong way to fix this.
Volker
(This used to be commit 9a04750dea)
by aliguori: NegTokenInit.mechListMIC is an Octet String.
Second: add a free_spnego_data function.
Both thanks to aliguori.
Volker
(This used to be commit 6c252440fb)
* use DsEnumerateDomainTrusts() instead of LDAP search.
wbinfo -m now lists all trusted downlevel domains and
all domains in the forest.
Thnigs to do:
o Look at Krb5 connection trusted domains
o make sure to initial the trusted domain cache as soon
as possible
(This used to be commit 0ab00ccaed)
maintain another tree then please apply!
On non-X86 machines out byte-order macros fails for one particular
value. If you asked for IVAL() of 0xFFFFFFFF and assigned it to a 64
bit quantity then you got a 63 bit number 0x7FFFFFFFFFFFFFFF rather
than the expected 0xFFFFFFFF. This is due to some rather bizarre and
obscure sign extension rules to do with unsigned chars and arithmetic
operators (basically if you | together two unsigned chars you get a
signed result!)
This affected a byte range lock using the large lockingX format and a
lock of offset 0 and length 0xFFFFFFFF. Microsoft Excel does one of
these locks when opening a .csv file. If the platform you run on does
not then handle locks of length 0x7FFFFFFFFFFFFFFF then the posix lock
fails and the client is given a lockingX failure. This causes the .csv
file to be trunated!!
(This used to be commit 886661c377)
to allow UNIX password change scripts to work correctly. This is safe as
the old password has been checked as correct before invoking this.
Jeremy.
(This used to be commit 1734d43eb5)
make sure we don't allow the creation of directories containing
wildcard characters. I've only put this in mkdir at the moment, but I
suspect this will apply to all places that can create new filenames.
We need to allow the opening of existing filenames that contain
wildcards, but not allow the creation of new ones.
Jeremy.
(This used to be commit 7f111e545d)
it's a different rc than KRB5_BAD_ENCTYPE (which exists on both MIT and
Heimdal). This will just make the debug show up at level 3 always.
Jeremy, you may want to revisit this, but it's probably not worth the hassle.
(This used to be commit 4ff322ccf9)
didn't behave exactly as I thought it did. If you create an item using a
PyFoo_FromBar function you must decrement the reference to that object
afterwards (or use Py_BuildValue).
(This used to be commit 2a960963fb)
No change to what is displayed has been made at this time. I do intend to
change the display order before 3.0.0 ships.
(This used to be commit de7d3063d9)
strdup_upper(). This function may fail - and we can just drop out of using
the cache in that case. (Rather than panicing).
This also should get us closer to supporting all of the weird 'longer/shorter'
on uppercase/lowercase.
Andrew Bartlett
(This used to be commit d4c9261725)
Remove strdup_upper/strdup_lower from their old file, now that they have
been moved to charcnv.c
Note that string_replace assumes that s is a pstring. (doco change only)
Andrew Bartlett
(This used to be commit 6c9056029b)
- Treat the NMB names in the 'session request' packet as 'ASCII'. This means
that we do not get invalid multibyte from the wire, even if we truncate
in the conversion. (Otherwise we panic when we try to strupper_m it).
- Remove acnv_uxu2(), as it was duplicated by push_ucs2_allocate()
- Remove acnv_dosu2(), as it is not used.
- In push_ucs2(), with the STR_UPPER flag, do the case conversion *after*
the UCS2 conversion, when it we know that the length can't change. Also
faster, as we don't need to do another 2 UCS2 conversions.
Andrew Bartlett
(This used to be commit 912035af11)
Server code *should* also work (I'll check shortly). May be the odd memory
leak. Problem was we (a) weren't setting signing on in the client krb5 sessionsetup
code (b) we need to ask for a subkey... (c). The client and server need to
ask for local and remote subkeys respectively.
Thanks to Paul Nelson @ Thursby for some sage advice on this :-).
Jeremy.
(This used to be commit 3f9e3b6070)
connection that set it up has been shut down.
(Also, pipes still connected, and reconnections to the same pipe (eg SAMR)
may continue to use that session key until their TCP/IP connection is shut
down)
Allow further testing by printing out the session key, and allowing it's input
into rpcclient.
Next step is automatic storage in a TDB.
Andrew Bartlett
(This used to be commit fa4d7be161)
discover names for the SAMR specific permissions that were previously unknown.
The existing constant names differ from what win2k calls them but since they
aren't heavily used in Samba at the moment I'll leave them as they are.
Jean-Baptiste's data is at:
http://ethereal.ntop.org/lists/ethereal-dev/200307/msg00314.html
(This used to be commit ae77e9e554)
due to w2k bug. I think this code is now working.... Need more testing of course
but works on all the obvious cases I can think of.
Jeremy.
(This used to be commit a6e537f661)
in oplock break state, change notify queue) we also push the MID onto
the deferred signing queue. Tomorrow I will test this with valgrind and
oplock tests.
Jeremy.
(This used to be commit 33a377f372)
to pstr_sprintf() and fstr_sprintf() to try to standardize.
lots of snprintf() calls were using len-1; some were using
len. At least this helps to be consistent.
(This used to be commit 9f835b85dd)
- free talloc context when cli_query_secdesc() fails
- dispose of cli_state when python cli_state_object is garbage collected
(This used to be commit 36052f95ff)
displaying pid_t, uid_t and gid_t values. This removes a whole lot of warnings
on some of the 64-bit build farm machines as well as help us out when 64-bit
uid/gid/pid values come along.
(This used to be commit f93528ba00)
Tested on a large combination of operating systems and versions.
Hopefully the build farm will find any remaining nasties if they
exist.
(This used to be commit 2e42fa3d72)
An extra message notifying that needed file didn't exist is displayed.
There's still a little catch with tdb backend, but it's better than it was,
from end-user's point of view.
This fixes#198
rafal
(This used to be commit b0be700605)
