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.
We now accept both \\server\tmp and \server\tmp. There are other places
where this might be necessary, but at least the functionality is a bit
easier now.
There are now ony 17 pstrings left in reply.c,
and these will be easy to remove (and I'll be
doing that shortly). Had to fix an interesting
bug in pull_ucs2_base_talloc() when a source
string is not null terminated :-).
Jeremy.
----------------------------------------------------------
In rare cases, Samba 3.0.25b shows directory contents at the wrong
position in the file tree when displaying a subdirectory of a DFS link.
The problem occurs whenever Windows XP asks for a DFS referral for a
subdirectory of a DFS link with a trailing backslash.
Windows does not do this very often, but we saw it several times per day
on our central DFS server.
smbd/msdfs.c, dfs_path_lookup() does the following with the requested
path:
- in line 390, the local copy 'localpath' is 'unix_convert'ed; the
trailing backslash is removed inside unix_convert
- in lines 417-20, 'dfspath' (another copy of the requested path) is
mangled another way without removing trailing backslashes
That's why the following loop (lines 435-461) that is meant to
synchronously cut off the last path component from both strings until it
comes to a DFS link, does not handle both strings the same. When the
original path ended with a backslash, 'canon_dfspath' has always one
component more than 'localpath', so that *consumedcntp gets too big in
line 446. This value is reported to the client.
----------------------------------------------------------
Bug #4860.
Jeremy.
why check_path_syntax should not be able to run in-line. The destination
pointer either walks side by side with the source pointer or is
decremented. So as far as I can see s>=d is true throughout the whole
routine.
Jeremy, I'm checking this only into 3_0 for now. Please review and ack
or directly merge this to 3_0_26.
Thanks,
Volker
"host msdfs = true" to be set in the [global] section
and allow Vista to see shares with "msdfs root = yes"
and "msdfs root = no" off the same server. Down
to an error message really :-).
Jeremy.
were not able to connect to the rewritten dfs code as
they set the dfs flag bit but then send local paths.
Now that our dfs code is a *lot* more robust in
detecting this sort of braindamage we can just
call into it directly on getting a DFS flag
and let the parser sort it out without having
to check it's actually connecting to a dfs
enabled share (I'm proud of this code :-).
Jeremy.
works from smbclient and Windows, and I am promising to
support and fix both client and server code moving forward.
Still need to test the RPC admin support but I haven't
changed that code.
Jeremy.
pathnames. When we're working out how much we've
consumed we need to backtrack by either a '/' or '\\'
component, as both are valid separators.
Jeremy.
Attached find a workaround that works for me. This is not the "correct" fix,
to me it seems our DFS referral marshalling is broken. Vista requests level 4,
we reply with level 2, and Vista seems not to like that. If we reply with
level 3 it seems more happy.
Needs more work!
Volker
This is actually an interesting case as it exposed bad code in our DFS redirect
module (that was where the bug was introduced).
Caused by our turning on dfsroot be default.
Jeremy.
* Remove "unknown" from dfs_Enum (samba4 dfs IDL updates to follow).
* When encountering an unsupported infolevel the rpc server must reply
with a dfs_info_0 structure and WERR_OK (observed from w2k3 when talking
to nt4).
Guenther
fact that check_path_syntax() will convert '\\' characters to '/'.
When POSIX pathnames have been selected this doesn't happen, so we
must look for the unaltered separator of '\\' instead of the modified '/'.
Stevef please check this with the CIFSFS MS-DFS code !
Jeremy
realloc can return NULL in one of two cases - (1) the realloc failed,
(2) realloc succeeded but the new size requested was zero, in which
case this is identical to a free() call.
The error paths dealing with these two cases should be different,
but mostly weren't. Secondly the standard idiom for dealing with
realloc when you know the new size is non-zero is the following :
tmp = realloc(p, size);
if (!tmp) {
SAFE_FREE(p);
return error;
} else {
p = tmp;
}
However, there were *many* *many* places in Samba where we were
using the old (broken) idiom of :
p = realloc(p, size)
if (!p) {
return error;
}
which will leak the memory pointed to by p on realloc fail.
This commit (hopefully) fixes all these cases by moving to
a standard idiom of :
p = SMB_REALLOC(p, size)
if (!p) {
return error;
}
Where if the realloc returns null due to the realloc failing
or size == 0 we *guarentee* that the storage pointed to by p
has been freed. This allows me to remove a lot of code that
was dealing with the standard (more verbose) method that required
a tmp pointer. This is almost always what you want. When a
realloc fails you never usually want the old memory, you
want to free it and get into your error processing asap.
For the 11 remaining cases where we really do need to keep the
old pointer I have invented the new macro SMB_REALLOC_KEEP_OLD_ON_ERROR,
which can be used as follows :
tmp = SMB_REALLOC_KEEP_OLD_ON_ERROR(p, size);
if (!tmp) {
SAFE_FREE(p);
return error;
} else {
p = tmp;
}
SMB_REALLOC_KEEP_OLD_ON_ERROR guarentees never to free the
pointer p, even on size == 0 or realloc fail. All this is
done by a hidden extra argument to Realloc(), BOOL free_old_on_error
which is set appropriately by the SMB_REALLOC and SMB_REALLOC_KEEP_OLD_ON_ERROR
macros (and their array counterparts).
It remains to be seen what this will do to our Coverity bug count :-).
Jeremy.
the name (must be abolute - start with /, must not end in /,
must have ./ and ../ removed). Of course for realpath resolved
paths this won't be the case but for others we need this name
to be canonicalized. This name is going into the sharemode db
for #3303 so needs to be in a normalized format.
Jeremy.
only tell at parse time from the wire if an incoming name
has wildcards or not. If it's a mangled name and we demangle
the demangled name may contain wildcard characters. Ensure
these are ignored.
Jeremy.
