search for: connection_struct

..., it is not very good idea to manage them with by-directional link list, for tracing link list will cause 'jump around the page' and 'jump around the cache' behavior. So, instead, I made 'array of pointer' with 128 pointers. ( I didn't change structure of `connection_struct', so they still contains next/prev pointers. It might be true that we no longer need it, or we can use it for another perpose ... like managing snum in some way better then simple search ). 3) I made entire "global" variable to single structure. do % patch -p0 < [The...

Problem is apparently with locking issues, disabled oplocks in the [general] section, and the problem actually got worse... Here's what happens: User-A part of group1, opens Excel file off of share, saves, exits... User-B (or even User-A for that matter) tries to re-open same file, get error stating it's locked and can only open for read-only access... Both users in the same group, and

...(*pp_smb_fname)->base_name); } ``` which redirects us to the `check_name` or the `check_name_with_privilege` function. which themselves respectively redirects to the `check_reduced_name` and `check_reduced_name_with_privilege` functions: source3/smbd/vfs.c: ``` NTSTATUS check_reduced_name(connection_struct *conn, const char *fname) [...] /* Common widelinks and symlinks checks. */ if (!allow_widelinks || !allow_symlinks) { [...] if (strncmp(conn_rootdir, resolved_name, rootdir_len) != 0) { DEBUG(2, ("check_reduced_name: Bad access "...

T.D.Lee@durham.ac.uk wrote: > > > But we'd like to go one step further and have the return code from preexec > be used to continue or abort the connection being established. The script > above would be modified to generate this return code. > > Looking through the source code (smbd/service.c), it doesn't seem to check > the return code (calling smbrun to do the

...t; > which redirects us to the `check_name` or the > `check_name_with_privilege` function. > which themselves respectively redirects to the `check_reduced_name` and > `check_reduced_name_with_privilege` functions: > > source3/smbd/vfs.c: > ``` > NTSTATUS check_reduced_name(connection_struct *conn, const char *fname) > [...] > /* Common widelinks and symlinks checks. */ > if (!allow_widelinks || !allow_symlinks) { > [...] > if (strncmp(conn_rootdir, resolved_name, > rootdir_len) != 0) { > DEBUG(2, ("check_reduced_nam...

..._OP_WRITE, SMB_VFS_OP_LSEEK, SMB_VFS_OP_SENDFILE, ... SMB_VFS_OP_LAST } vfs_op_type; This struct contains the function and handle pointers for all operations. struct vfs_ops { struct vfs_fn_pointers { ... /* File operations */ int (*open)(struct vfs_handle_struct *handle, struct connection_struct *conn, const char *fname, int flags, mode_t mode); int (*close)(struct vfs_handle_struct *handle, struct files_struct *fsp, int fd); ssize_t (*read)(struct vfs_handle_struct *handle, struct files_struct *fsp, int fd, void *data, size_t n); ssize_t (*write)(struct vfs_handle_struct *h...

...++-------------------- > 1 file changed, 64 insertions(+), 39 deletions(-) > > diff --git a/source3/smbd/open.c b/source3/smbd/open.c > index baebd7c..bc676f9 100644 > --- a/source3/smbd/open.c > +++ b/source3/smbd/open.c > @@ -3397,6 +3397,7 @@ static NTSTATUS open_directory(connection_struct *conn, > struct timespec mtimespec; > int info = 0; > bool ok; > + bool need_lstat = false; > > if (is_ntfs_stream_smb_fname(smb_dname)) { > DEBUG(2, ("open_directory: %s is a stream name!\n", > @@ -3502,25 +3503,9 @@ static NTSTATUS open_directory(con...

Is behavior of ACLs under Samba 3.3.0 (Sernet) completely different from that under version 3.2.7? The release notes only talks about some "fixes". I installed version 3.3.0 and got completely different result with the same filesystem and the exact same samba configuration. The ACLs behaved strangely and appeared very different under Windows ACL editor. Users were now unable to

...samba-3.0.23d/source/smbd/reply.c 2006-06-23 08:16:49.000000000 -0500 +++ samba/source/smbd/reply.c 2007-01-30 15:00:45.000000000 -0600 @@ -1865,7 +1865,7 @@ Check if a user is allowed to delete a file. ********************************************************************/ -NTSTATUS can_delete(connection_struct *conn, char *fname, uint32 dirtype, BOOL bad_path, BOOL check_is_at_open) +NTSTATUS can_delete(connection_struct *conn, char *fname, uint32 dirtype, BOOL bad_path, BOOL check_is_at_open, BOOL can_defer) { SMB_STRUCT_STAT sbuf; uint32 fattr; @@ -1938,7 +1938,7 @@ FILE_OPEN, 0,...

...samba-3.0.23d/source/smbd/reply.c 2006-06-23 08:16:49.000000000 -0500 +++ samba/source/smbd/reply.c 2007-01-30 15:00:45.000000000 -0600 @@ -1865,7 +1865,7 @@ Check if a user is allowed to delete a file. ********************************************************************/ -NTSTATUS can_delete(connection_struct *conn, char *fname, uint32 dirtype, BOOL bad_path, BOOL check_is_at_open) +NTSTATUS can_delete(connection_struct *conn, char *fname, uint32 dirtype, BOOL bad_path, BOOL check_is_at_open, BOOL can_defer) { SMB_STRUCT_STAT sbuf; uint32 fattr; @@ -1938,7 +1938,7 @@ FILE_OPEN, 0,...

...pool-usage' shows: global talloc allocations in pid: 13843 name chunks bytes ---------------------------------------- -------- -------- msg_pool_usage 1 159 lp_talloc 2 2 connection_struct 1 1844 pipe spoolss 0x922dfb0 0 0 pipe spoolss 0x83f4380 0 0 connection_struct 1 1844 connection_struct 1 1844 passdb internal SAM_AC...

I have "hide dot files" on, but I have noticed that while files with a single dot are hidden, files with multiple dots are not: ".foo" - hidden "...foo" - not hidden Is this by design? *nix certainly hide such files by default... -- Barry -- http://barrkel.blogspot.com/

...oo' in passdb. [2002/07/12 09:28:18, 1] smbd/reply.c:reply_sesssetup_and_X(989) Rejecting user 'foo': authentication failed I am using samba version 2.2.4 in a RedHat Linux box. Browsing the source at smbd/reply.c I see no possibility to include source IP address in the log message (connection_struct *conn is NULL at this point). Neither it is possible at smbd/password.c. I think this feature is pretty interesting to determine where some kinds of attacks are coming from (especially the ones trying to guess privileged accounts and their passwords in a samba server). Any comments will be much a...

Hi, I'm using samba version samba-4.1.6+dfsg included in last ubuntu LTS version. Here is my smb.conf file: [global] # configuration du serveur netbios name = scribe workgroup = dompedago server string = scribe preferred master = yes domain logons = yes security = user ldap passwd sync = yes passdb backend =

We looked at this some more, and Andrew seemed to understand and wrote the attached patch. > > We got the logs by forcing smbd to run with -d10 by patching > file_server/fileserver.c. The issue appears to be in this call: 3638 /* Ensure there was no race condition. */ 3639 if (!check_same_stat(&smb_dname->st, &fsp->fsp_name->st)) { 3640

...rojects at the moment. VL: One particularly confusing part of Samba 3.x is the way we represent the "current user" information. We have multiple places where we represent the current credentials. For example there is the "current_user" global variable. In "connection_struct" we have sets of user-ids, there is the "struct user_struct" and so on. I want to replace all those places where we store and use the current user information by passing around the central structure that falls out of authentication: The "auth_serversupplied_info...

...rojects at the moment. VL: One particularly confusing part of Samba 3.x is the way we represent the "current user" information. We have multiple places where we represent the current credentials. For example there is the "current_user" global variable. In "connection_struct" we have sets of user-ids, there is the "struct user_struct" and so on. I want to replace all those places where we store and use the current user information by passing around the central structure that falls out of authentication: The "auth_serversupplied_info...

...> > which redirects us to the `check_name` or the > `check_name_with_privilege` function. > which themselves respectively redirects to the `check_reduced_name` and > `check_reduced_name_with_privilege` functions: > > source3/smbd/vfs.c: > ``` > NTSTATUS check_reduced_name(connection_struct *conn, const char *fname) > [...] >     /* Common widelinks and symlinks checks. */ >     if (!allow_widelinks || !allow_symlinks) { > [...] >         if (strncmp(conn_rootdir, resolved_name, >                 rootdir_len) != 0) { >             DEBUG(2, ("check_reduced_nam...

There was another tech coordinator that asked the following on our list: > Hey all, I was wondering if anyone knew of a way to give students read/write permissions to a certain drive letter without giving them the ability to delete other (or thier own) files. Currently I have chmod 1770 permissions (read write, but only owner can delete) and one owner (the teacher) for all the files, but

Hi all, On Thu, Apr 16, 2015 at 03:00:49PM -0700, Jeremy Allison wrote: > On Thu, Apr 16, 2015 at 08:42:48PM +0200, Ervin Heged?s wrote: > > Dear Samba users, > > > > here is an Ubuntu 14.04, with Samba 4 (4.1.6), and LDAP (slapd > > 2.4.31). The config came from a previous system (Debian Squeezy), > > which had been crashed (HW error - on this new machine,