summaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2009-03-19lib/tsocket: add tsocket_guide.txt section about the async _send/recv() ↵Stefan Metzmacher1-0/+142
helper functions metze
2009-03-19lib/tsocket: add tsocket_readv_send/recv()Stefan Metzmacher3-1/+235
metze
2009-03-19lib/tsocket: add tsocket_writev_queue_send/recv()Stefan Metzmacher2-0/+126
metze
2009-03-19lib/tsocket: add tsocket_writev_send/recv()Stefan Metzmacher3-1/+195
metze
2009-03-19lib/tsocket: add tsocket_connect_send/recv()Stefan Metzmacher3-1/+129
metze
2009-03-19lib/tsocket: add tsocket_sendto_queue_send/recv()Stefan Metzmacher2-0/+133
metze
2009-03-19lib/tsocket: add tsocket_sendto_send/recv()Stefan Metzmacher3-1/+145
metze
2009-03-19lib/tsocket: add tsocket_recvfrom_send/recv()Stefan Metzmacher3-2/+178
metze
2009-03-19lib/tsocket: add an implementation for bsd style sockets.Stefan Metzmacher3-1/+1169
metze
2009-03-19lib/tsocket: add a tsocket_guide.txt with a description of the designStefan Metzmacher1-0/+282
metze
2009-03-19lib/tsocket: add some useful helper functionsStefan Metzmacher3-1/+183
metze
2009-03-19lib/tsocket: add generic socket abstraction layerStefan Metzmacher6-0/+521
This will replace source4/lib/socket/. metze
2009-03-19fix build on old Heimdal based systemsBjörn Jacke1-5/+3
Signed-off-by: Günther Deschner <gd@samba.org>
2009-03-19tevent: fix the nesting logicStefan Metzmacher1-7/+7
Only tevent_loop_once and tevent_loop_until() should care about the nesting level. This fixes the samba3 printing code where we use tevent_loop_wait() and don't allow nested events. We still call the nesting hook for all levels, we need to decide if we really want this... metze
2009-03-19s3:libsmb: always create bytes array in cli_trans codeStefan Metzmacher1-5/+14
Otherwise we return NO_MEMORY without a reason for fragmented trans requests, as talloc_append_blob() returns buf if we append a 0 length blob. When we pass buf = NULL we'll get back NULL and then assume NO_MEMORY... metze
2009-03-19s3:libsmb: fix smb signing for fragmented trans/trans2/nttrans requestsStefan Metzmacher1-23/+7
Before we send the secondary requests we need to remove the old mid=>seqnum mapping and reset cli->mid and make the new mid=>seqnum mapping "persistent". The bug we had in cli_send_trans was this: The first cli_send_smb() incremented cli->mid and the secondary requests used the incremented mid, but as cli->outbuf still had the correct mid, we send the correct mid to the server. The real problem was that the cli_send_smb() function stored the seqnum under the wrong mid. cli_send_nttrans() was totally broken and now follows the same logic as cli_send_trans(). The good thing is that in practice the problem is unlikely to happen, because max_xmit is large enough to avoid secondary requests. metze
2009-03-19s3:lib/util_sock: use sys_recv() instead of sys_read() on socketsStefan Metzmacher1-2/+2
This ways the pcap support in socket wrapper sees the received data. metze
2009-03-19fix some typosBjörn Jacke1-3/+3
Signed-off-by: Michael Adam <obnox@samba.org>
2009-03-18Modify simple POSIX open test to use filenames containing a ':'Jeremy Allison1-2/+2
character. Should stop regressions of bug #6196. Jeremy.
2009-03-18Allow DFS client paths to work when POSIX pathnames have beenJeremy Allison1-2/+12
selected (we need to path in pathname /that/look/like/this). Jeremy.
2009-03-18Merge branch 'master' of ssh://jra@git.samba.org/data/git/sambaJeremy Allison2-1/+79
2009-03-18Fix bug #6196 - Unable to serve files with colons to Linux CIFS/VFS clientJeremy Allison1-2/+7
Looks like the pathname parsing for POSIX paths got broken when the code for doing Windows streams parsing got added. Jeremy.
2009-03-19use the tevent nesting code to avoid the uid problem in the VFSAndrew Tridgell1-0/+66
backend The vfs_unixuid module changes the uid of the process when executing operations on behalf of the user. Within the VFS backend we may rely on semi-async calls, such as winbind calls, which will call the event loop again. To cope with this we need to ensure that while inside those calls we revert the uid to root, then revert back to the connected user when we have finished with the semi-async calls.
2009-03-19fixed a logic bug in the tevent nesting codeAndrew Tridgell1-1/+13
The event nesting code never triggered as nesting.level was never greater than 1. The main event loop needs to increase the nesting level by 1. I also added a paranoia check to the nesting setup call. The API as currently written cannot support multiple nesting hooks, so we need to abort if multiple hooks are tried.
2009-03-18s3 onefs: Correctly error out when the read returns EOFTim Prouty1-5/+9
Also add some more debugging.
2009-03-18Fix bug #6195 - Migrating from 3.0.x to 3.3.x can fail to update passdb.tdb ↵Jeremy Allison2-19/+206
correctly. This is a really nasty one to fix as in order to successfully update the passdb.tdb we must do the equivalent of a tdbbackup to move to the new hash values before we do the upgrade. Jeremy.
2009-03-18s3: remove POLICY_HND.Günther Deschner43-376/+360
Guenther
2009-03-18s3-rpc_parse: remove some unused parsing code.Günther Deschner2-351/+0
Guenther
2009-03-18s3-spoolss: fix _spoolss_GetPrinterData printerserver handle query error code.Günther Deschner1-1/+1
When _spoolss_GetPrinterData receives a query on a printserver handle for a value that we have not stored or do not provide, we need to return WERR_INVALID_PARAM, not WERR_BADFILE. Tested with w2k and w2k3 servers. Found by torture test. Guenther
2009-03-18s3-spoolss: fix _spoolss_EnumPrinterDataEx error path.Günther Deschner1-0/+4
When a windows clients queries the "" key, we need to make sure to return with the appropriate error (WERR_INVALID_PARAM in that case), and not fall through to the buffer size handling macros. Found by torture test. Guenther
2009-03-18s3-rpc_parse: remove unused BUFFER5 and UNISTR3.Günther Deschner5-183/+0
Guenther
2009-03-18s3-rpc_client: remove unused CLI_DO_RPC_WERR macro.Günther Deschner1-30/+0
Guenther
2009-03-18s4 libcli: Add back Longhorn beta dialect as a constantZach Loafman2-4/+4
Also tweak how the dialect array and count is initialized.
2009-03-18s3-smbcontrol: use correct PRINTER_NOTIFY flags.Günther Deschner1-3/+3
Guenther
2009-03-18spoolss: add my copyright.Günther Deschner4-1/+4
Guenther
2009-03-18s3-spoolss: remove include/rpc_spoolss.h.Günther Deschner2-168/+0
Guenther
2009-03-18s3-spoolss: remove rpc_server/srv_spoolss.c alltogether.Günther Deschner4-601/+5
Guenther
2009-03-18s3-spoolss: remove rpc_parse/parse_spoolss.c alltogether.Günther Deschner3-33/+1
Good-Bye, last hand-marshalled rpc functions, rest in peace. Guenther
2009-03-18s3-spoolss: remove old spoolss_EnumPrinterDataEx.Günther Deschner3-294/+0
Guenther
2009-03-18s3-spoolss: use pidl for _spoolss_EnumPrinterDataEx.Günther Deschner3-115/+60
Please note that this has been the last call in samba3 that was using hand-marshalled rpc. With this commit all named pipe rpc services in samba3 have now fully moved to pidl generated code :-) Guenther
2009-03-18s3-spoolss: use printer and job notify enums provided by idl.Günther Deschner3-109/+109
Guenther
2009-03-18s3-spoolss: fix spoolss server after spoolss_Field changes.Günther Deschner2-12/+12
Guenther
2009-03-18s3-rpcclient: fix spoolss notify test after spoolss_Field changes.Günther Deschner1-4/+4
Guenther
2009-03-18s4-smbtorture: fix spoolss notify test after spoolss_Field changes.Günther Deschner1-4/+4
Guenther
2009-03-18spoolss: add custom ndr_print_spoolss_Field.Günther Deschner2-0/+23
Guenther
2009-03-18s3: re-run make samba3-idl.Günther Deschner3-90/+288
Guenther
2009-03-18spoolss: make spoolss_Field a nodiscriminant union of 2 sets of notify flag ↵Günther Deschner1-30/+65
enums. Guenther
2009-03-18s3-rpc_parse: move prs_uint64 to rpc_parse/parse_prs.c.Günther Deschner3-25/+25
Guenther
2009-03-18s3-spoolss: remove custom syntax_spoolss and use the syntax defined in IDL.Günther Deschner8-72/+60
Guenther
2009-03-18s3-printing: use marshall/unmarshall_sec_desc_buf in sec_desc_upg_fn().Günther Deschner1-26/+5
Guenther