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 is intermediate fix as discussed with Jeremy until we move check_path_syntax() to UCS2 internally where all
ambiguity is resolved. Please add other encodings into charcnv.c with such property.
(This used to be commit c8c66337494c63173a6aba4286a843875302da63)
This is intermediate fix as discussed with Jeremy until we move check_path_syntax() to UCS2 internally where all
ambiguity is resolved. Please add other encodings into charcnv.c with such property.'
'
(This used to be commit 2c404f6ba988d68c6f44df9409c0de319553de10)
names in nmbd. Allows conversion from dos codepage mb strings (ie. SJIS) to
expand to utf8 size on read.
Jeremy.
(This used to be commit b9da3e941d4233c2d3931f32767f3e51b214033e)
names in nmbd. Allows conversion from dos codepage mb strings (ie. SJIS) to
expand to utf8 size on read.
Jeremy.
(This used to be commit 834d816caf9cd6318da00febde50d9233469dac2)
to winbindd. idmap_allocate_rid wants information about whether this will be a
user or a group, I did not export this to the winbind interface.
The reason for idmap to get that info is to keep consistent with the
algorithmic convention to alloc only even rids for users and odd rids for
groups. I'm not fully convinced that this really gains us anything. Any real
good arguments?
Volker
(This used to be commit 7f62cf933cad69799204bfdc773e08ff0dde0b20)
much of your winbindd_passdb, users are currently not provided by that, only
aliases. Currently the code to maintain that stuff is not yet in, this will be
next, see my next posting to samba-technical.
Volker
(This used to be commit 9e0fb457ba77a55f8271b6acc91a07f0a8df3760)
enum_aliases to passdb. create and delete are not yet filled, you have to do
this with net groupmap and net idmap restore.
Volker
(This used to be commit f44bf63d1143ddf6a1a2fdc15a301e0dc0840d50)
Don't use nstrings to hold workgroup and netbios names. The problem with them is that MB netbios
and workgroup names in unix charset (particularly utf8) may be up to 3x bigger than the name
when represented in dos charset (ie. cp932). So go back to using fstrings for these but
translate into nstrings (ie. 16 byte length values) for transport on the wire.
Jeremy.
(This used to be commit 128dec9ae68fd357550de2649d54056ca4fc65cf)
Don't use nstrings to hold workgroup and netbios names. The problem with them is that MB netbios
and workgroup names in unix charset (particularly utf8) may be up to 3x bigger than the name
when represented in dos charset (ie. cp932). So go back to using fstrings for these but
translate into nstrings (ie. 16 byte length values) for transport on the wire.
Jeremy.
(This used to be commit b4ea493599ab414f7828b83f40a5a8b43479ff64)
fix up netbios names with mb strings. Includes reformat of libsmb/nmblib.c
so it's readable.
Jeremy.
(This used to be commit beca3829d1140bf7c2e394993a02dff1f2cf7bbd)
fix up netbios names with mb strings. Includes reformat of libsmb/nmblib.c
so it's readable.
Jeremy.
(This used to be commit 966e49a48c352563cdd7f75fe2768f2d6612ec7e)
there are multiple "."'s in the name.
This code is protected with an #ifdef TRUNCATE_NETBIOS_NAME and this
is #define'd to 1 directly above. Should we also get rid of the #ifdef?
(This used to be commit 0375dace248eb3dc660fa2bca2808552e502b3f7)
This is an odd corner case having to do with <1C> group names (DC names).
<1C> group names are called (by MS) "Internet Special Groups", which means
that Microsoft's WINS servers will attempt to handle these names in
something approximating an RFC1001/1002 compliant manner.
The problem being reported here is this: If an initial registration sent
by one of the group members is lost (network error, whatever) then
subsequent refreshes from that particular machine will also fail. This
only happens if the name is already active (because of other group
members).
In most cases, we (and MS) handle refreshes as registrations if the name
is not in the database. In this situation, we missed the corner case.
Bert's fix adds an 'else if' that catches the situation in which a refresh
for an unlisted IP of an active <1C> group name is received. The refresh
is simply handled as a registration when this happens.
[Note: Committing in HEAD. I am writing some tools to do registrations
and refreshes so that I can test this. I don't have an NT Domain
for testing so I'm going to have to fudge. This fix is small (I
cleaned up some Debug messages and comments in addition to the fix)
and if you want to test it with 3.0.x you can just copy the HEAD
version into your 3.0.x tree and recompile.]
(This used to be commit 8e52439f4ae9de84107e2d0e9f33f0d4e3c5945a)
It appears that we pass filename through resolve_wildcards() as pstring and use fstring temporary buffer there.
As result, a long filename in unix charset (UTF-8 by default) can easily expand over 255 bytes while
Windows is able to send to us such names (e.g. Japanese name of ~190 mb chars) which we unable to process through
this small fstring buffer. Tested with W2K and smbclient, Japanese and Cyrillics.
(This used to be commit 83dac6571f99b854ac607d4313cc3f742c1fae2e)
It appears that we pass filename through resolve_wildcards() as pstring and use fstring temporary buffer there.
As result, a long filename in unix charset (UTF-8 by default) can easily expand over 255 bytes while
Windows is able to send to us such names (e.g. Japanese name of ~190 mb chars) which we unable to process through
this small fstring buffer. Tested with W2K and smbclient, Japanese and Cyrillics.
(This used to be commit cc70a548f3c48a9503fd54d01896cc2499f5bbcd)
