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.
Right now, the nss_info backends are tied to the idmap backends (which is wrong
IMHO). In the domain child we don't load the idmap backend anymore, so we don't
have the nss info modules. This needs fixing properly.
Casting those variables will lead to sscanf believing that it sees pointers to
unsigned longs. These might be 64 bit long, thus sscanf will overwrite memory
it should not overwrite. Assigning the vars later is okay, there we get
automatic type conversion. C can be nasty ...
Christian, please check!
as the autorid module relies on a stable minimum uid/gid value
and rangesize, it now saves the values used at first successful start
and refuses to work if these values get changed in smb.conf later.
Changing the values after the first mapping was done will result
in unpredictable behaviour.
Another check covers the maximum uid value. If this gets decreased
later and domain range mappings already exist that would result
in uid values higher than the new uid value, initialization will
be aborted
These variables, of type struct auth_serversupplied_info were poorly
named when added into 2001, and in good consistant practice, this has
extended all over the codebase in the years since.
The structure is also not ideal for it's current purpose. Originally
intended to convey the results of the authentication modules, it
really describes all the essential attributes of a session. This
rename will reduce the volume of a future patch to replaced these with
a struct auth_session_info, with auth_serversupplied_info confined to
the lower levels of the auth subsystem, and then eliminated.
(The new structure will be the output of create_local_token(), and the
change in struct definition will ensure that this is always run, populating
local groups and privileges).
Andrew Bartlett
Signed-off-by: Stefan Metzmacher <metze@samba.org>
this brings the s3 waf build much closer to the proposed s3build top
level build, using the same bld.SAMBA3_*() rules
There are a few renames of subsystems in here, with a 3 suffix where
it would create a conflict.
Pair-Programmed-With: Andrew Bartlett <abartlet@samba.org>
The benefit of this that it makes us more robust to secure channel resets
triggered from tools outside the winbind process. Long term we need to have a
shared tdb secure channel store though as well.
Guenther
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Autobuild-User: Stefan Metzmacher <metze@samba.org>
Autobuild-Date: Fri Feb 4 18:11:04 CET 2011 on sn-devel-104
If winbindd connects to a domain controller it doesn't establish the lsa
connection over ncacn_ip_tcp direct. This happens only on demand.
If someone does a 'net rpc testjoin' and then a
wbinfo -n DOMAIN\\administrator, we'll get DCERPC faults with
ACCESS_DENIED/SEC_PKG_ERROR, because winbindd's in memory copy
of the schannel session key is invalidated.
This problem can also happen on other calls, but the
lookup_names/sids calls on thet lsa ncacn_ip_tcp connection
are the most important ones.
The long term fix is to store the schannel client state in a
tdb, but for now it's enough to catch the error and invalidate
the all connections to the dc and reestablish the schannel
session key.
The fix for bug 7568 (commit be396411a4e1f3a174f8a44b6c062d834135e70a)
made this worse, as it assumes winbindd's in memory session key is
always the current one.
metze
