summaryrefslogtreecommitdiff
path: root/source3/smbd
AgeCommit message (Collapse)AuthorFilesLines
2011-02-02Based on a conversation with Volker, refactor some of the oplock code to ↵Jeremy Allison1-71/+160
make it comprehensible. delay_for_oplocks() did 4 things. 1). Validation of existing oplock types. 2). Check for compatibility with batch oplocks (pass 1). 3). Check for compatibility with exclusive oplocks (pass 2). 4). Set the correct oplock type from the requested value. Refactor into 4 separate functions: 1). find_oplock_types() - does validation of oplock types and returns pointers to specific values. 2). delay_for_batch_oplocks() - the pass 1 phase above. 3). delay_for_exclusive_oplocks() - the pass 2 phase above 4). grant_fsp_oplock_type() - Set the correct oplock type from the requested value. Now separated out this code should be much easier to understand and modify. This also fixes an erroneous SMB_ASSERT which was hidden by the previous complexity of the single delay_for_oplocks() code. Jeremy. Autobuild-User: Jeremy Allison <jra@samba.org> Autobuild-Date: Wed Feb 2 01:52:21 CET 2011 on sn-devel-104
2011-02-01s3-lanman: prefer dcerpc_samr_X functions.Günther Deschner1-83/+237
Guenther Autobuild-User: Günther Deschner <gd@samba.org> Autobuild-Date: Tue Feb 1 16:45:10 CET 2011 on sn-devel-104
2011-02-01Ensure we send the direct levelII oplock break to the correct fid.Pavel Shilovsky1-1/+10
Autobuild-User: Jeremy Allison <jra@samba.org> Autobuild-Date: Tue Feb 1 00:39:56 CET 2011 on sn-devel-104
2011-01-31Fix bug #7928 - Samba problems with kernel oplocks option set to "no"Pavel Shilovsky2-8/+32
We should not grant levelII oplocks on a file with existing byte range locks.
2011-01-31s3:smbd: let smbd_server_connection_loop_once() check for select errorsStefan Metzmacher1-0/+4
metze
2011-01-31Revert "s3: Fix starving the echo responder"Stefan Metzmacher1-17/+0
This reverts commit d5cf6482ed0cd9a11448ca04944b2e01200a7c89. I'll add a more generic fix for this problem. metze
2011-01-31Revert "s3:events: Call all ready fd event handlers on each iteration of the ↵Stefan Metzmacher1-18/+21
main loop" This reverts commit 455fccf86b6544cd17a2571c63a88f8aebff3f74. I'll add a more generic fix for this problem. metze
2011-01-28We need to reply to SMB2_GETINFO_FILE with a class of ↵Jeremy Allison1-1/+32
SMB2_FILE_STANDARD_INFO on an IPC$ share. Otherwise a Win7 client issues this request twice (2xroundtrips) if we return NOT_SUPPORTED. We do the same for SMB1 in call_trans2qpipeinfo() Jeremy. Autobuild-User: Jeremy Allison <jra@samba.org> Autobuild-Date: Fri Jan 28 22:53:18 CET 2011 on sn-devel-104
2011-01-27Change "security=share" to downgrade to SMB1 from SMB2, rather than trying ↵Jeremy Allison1-8/+0
to fake it.
2011-01-27Looking into printer driver issues, I ran across some peculiarities inDavid Disseldorp1-8/+11
copy_file(): - Firstly, if the source file is zero bytes, NT_STATUS_DISK_FULL is returned. - Secondly, the conditional lseek is confusing. It fires when OPENX_FILE_EXISTS_OPEN is set and I can't see why the lseek is necessary in this case. - Finally, the lseek error path also results in NT_STATUS_DISK_FULL. Proposed fix for first and third point below. Autobuild-User: Jeremy Allison <jra@samba.org> Autobuild-Date: Thu Jan 27 00:59:56 CET 2011 on sn-devel-104
2011-01-26Allow "security = share" with SMB2. We already handle this in smb2/smb2_server.cJeremy Allison1-1/+0
Autobuild-User: Jeremy Allison <jra@samba.org> Autobuild-Date: Wed Jan 26 20:37:29 CET 2011 on sn-devel-104
2011-01-25Fix bug #7863 - Unlink may unlink wrong file when hardlinks are involved.Jeremy Allison5-36/+78
Do this by keeping a linked list of delete on close tokens, one for each filename that identifies a path to the dev/inode. Use the jenkins hash of the pathname to identify the correct token.
2011-01-25Add uint32_t name_hash argument (currently unused) to get_file_infos().Jeremy Allison7-12/+13
Will be used when we store more than one delete on close token.
2011-01-25Add name_hash into the share mode entry struct (as yet only use for renames ↵Jeremy Allison2-1/+10
to identify a specific path).
2011-01-25Add name_hash to files_struct. Set within fsp_set_smb_fname().Jeremy Allison1-1/+32
2011-01-21s3-lanman: remove trailing whitespace.Günther Deschner1-30/+30
Guenther
2011-01-21s3:smbd: use dcerpc_spoolss_X() functionsStefan Metzmacher2-32/+50
metze Signed-off-by: Andreas Schneider <asn@samba.org>
2011-01-21s3: Fix a deadlock between smbd and ctdbdVolker Lendecke1-4/+18
Do the notification after we released the share mode lock. Inside notify_fname we take out another tdb lock. With ctdb also accessing our databases, this can lead to deadlocks. Putting this notify after the TALLOC_FREE(lck) above we avoid locking two records simultaneously. Notifies are async and informational only, so calling the notify_fname without holding the share mode lock should not do any harm. Autobuild-User: Volker Lendecke <vlendec@samba.org> Autobuild-Date: Fri Jan 21 12:29:21 CET 2011 on sn-devel-104
2011-01-21Remove all uses of "./" in pathnames - make canonical. This will become ↵Jeremy Allison1-50/+78
important when we need to guarantee canonical names for hashing. Autobuild-User: Jeremy Allison <jra@samba.org> Autobuild-Date: Fri Jan 21 02:59:56 CET 2011 on sn-devel-104
2011-01-20Ensure we don't use "./" in findfirst pathnames. Use a directory open of "." ↵Jeremy Allison1-4/+8
instead.
2011-01-20Make processing of incoming stream rename paths common between reply_mv and ↵Jeremy Allison2-4/+47
ntrename. Ensure we don't depend on "./" in the streams module.
2011-01-20s3:smbd: use anonymous_shared_free() for shared memory signing stateStefan Metzmacher1-0/+7
metze Autobuild-User: Stefan Metzmacher <metze@samba.org> Autobuild-Date: Thu Jan 20 07:48:29 CET 2011 on sn-devel-104
2011-01-20lib/util: s/allocate_anonymous_shared/anonymous_shared_allocate/Stefan Metzmacher1-1/+1
metze
2011-01-19s3-smbd: Fixed a possible null pointer dereference.Andreas Schneider1-1/+3
2011-01-17s3: Fix a C++ warningVolker Lendecke1-1/+1
2011-01-14s3: Fix bug 7917: Yet another bug in chain_replyVolker Lendecke1-2/+2
Found by Michael Hanscho <samba@micha.priv.at> with a WinCE client. Autobuild-User: Volker Lendecke <vlendec@samba.org> Autobuild-Date: Fri Jan 14 17:42:05 CET 2011 on sn-devel-104
2011-01-13s3-lanman: prefer dcerpc_srvsvc_X functions.Günther Deschner1-3/+9
Guenther Signed-off-by: Andreas Schneider <asn@samba.org>
2011-01-07s3-printing: Initiate pcap reload from parent smbdDavid Disseldorp5-40/+65
Since commit 7022554, smbds share a printcap cache (printer_list.tdb), therefore ordering of events between smbd processes is important when updating printcap cache information. Consider the following two process example: 1) smbd1 receives HUP or printcap cache time expiry 2) smbd1 checks whether pcap needs refresh, it does 3) smbd1 marks pcap as refreshed 4) smbd1 forks child1 to obtain cups printer info 5) smbd2 receives HUP or printcap cache time expiry 6) smbd2 checks whether pcap needs refresh, it does not (due to step 3) 7) smbd2 reloads printer shares prior to child1 completion (stale pcap) 8) child1 completion, pcap cache (printer_list.tdb) is updated by smbd1 9) smbd1 reloads printer shares based on new pcap information In this case both smbd1 and smbd2 are reliant on the pcap update performed on child1 completion. The prior commit "reload shares after pcap cache fill" ensures that smbd1 only reloads printer shares following pcap update, however smbd2 continues to present shares based on stale pcap data. This commit addresses the above problem by driving pcap cache and printer share updates from the parent smbd process. 1) smbd0 (parent) receives a HUP or printcap cache time expiry 2) smbd0 forks child0 to obtain cups printer info 3) child0 completion, pcap cache (printer_list.tdb) is updated by smbd0 4) smbd0 reloads printer shares 5) smbd0 notifies child smbds of pcap update via message_send_all() 6) child smbds read fresh pcap data and reload printer shares This architecture has the additional advantage that only a single process (the parent smbd) requests printer information from the printcap backend. Use time_mono in housekeeping functions As suggested by Björn Jacke.
2011-01-07s3-printing: reload shares after pcap cache fillDavid Disseldorp3-7/+12
Since commit eada8f8a, updates to the cups pcap cache are performed asynchronously - cups_cache_reload() forks a child process to request cups printer information and notify the parent smbd on completion. Currently printer shares are reloaded immediately following the call to cups_cache_reload(), this occurs prior to smbd receiving new cups pcap information from the child process. Such behaviour can result in stale print shares as outlined in bug 7836. This fix ensures print shares are only reloaded after new pcap data has been received. Pair-Programmed-With: Lars Müller <lars@samba.org>
2011-01-04s3-rpcecho: Only register rpcecho in the developer build.Andreas Schneider2-0/+4
Autobuild-User: Andreas Schneider <asn@samba.org> Autobuild-Date: Tue Jan 4 18:56:38 CET 2011 on sn-devel-104
2011-01-04s3-smbd: Call the rpc service shutdown functions.Andreas Schneider1-0/+34
2011-01-04s3-smbd: Call all the rpc services in the right order.Andreas Schneider1-5/+41
2010-12-29Fix bug #7892 - open_file_fchmod() leaves a stale lock.Jeremy Allison3-45/+10
Autobuild-User: Jeremy Allison <jra@samba.org> Autobuild-Date: Wed Dec 29 02:15:23 CET 2010 on sn-devel-104
2010-12-23All calls to event_add_to_select_args() call GetTimeOfDay() andJeremy Allison1-7/+2
pass this in as the &now parameter. Push this call inside of event_add_to_select_args() to the correct point so it doesn't get called unless needed. Jeremy. Autobuild-User: Jeremy Allison <jra@samba.org> Autobuild-Date: Thu Dec 23 01:08:11 CET 2010 on sn-devel-104
2010-12-21My algorithm for determining whan an incoming sequence number can be allowed ↵Jeremy Allison1-6/+7
is incorrect. (I based it on the text in MS-SMB2, silly me :-). Fix it so incoming sequence numbers can range over the entire allowable bitmap range. This fixes a repeatable disconnect against Win7. Jeremy.
2010-12-21Keep track of the sparse status of an open file handle. Allows bypass ofJeremy Allison3-1/+13
strict allocation on sparse files. Files opened as POSIX opens are always sparse. Autobuild-User: Jeremy Allison <jra@samba.org> Autobuild-Date: Tue Dec 21 04:12:22 CET 2010 on sn-devel-104
2010-12-21Added call out to a Linux-compatible fallocate() when we need to extend a fileJeremy Allison1-1/+12
allocation extent without changing end-of-file size. Autobuild-User: Jeremy Allison <jra@samba.org> Autobuild-Date: Tue Dec 21 02:41:24 CET 2010 on sn-devel-104
2010-12-20CREATE in a compound CREATE/NOTIFY sequence was being passed through ↵Ken Harris1-17/+17
set_operation_credits() twice (ultimately perhaps because of bug 7331 involving this compound sequence and the need to be ready for any incoming CANCEL of the NOTIFY). This had the server thinking it had granted more credit than it actually had, which lead to zero-credits being granted in interim NOTIFY responses. Autobuild-User: Jeremy Allison <jra@samba.org> Autobuild-Date: Mon Dec 20 20:59:55 CET 2010 on sn-devel-104
2010-12-20From metze's work on sparse attributes. FILE_ATTRIBUTE_SPARSE is valid on ↵Jeremy Allison1-1/+2
get but not on set. Autobuild-User: Jeremy Allison <jra@samba.org> Autobuild-Date: Mon Dec 20 20:11:22 CET 2010 on sn-devel-104
2010-12-19s3: Reply correctly to FSCTL_IS_VOLUME_DIRTYVolker Lendecke1-0/+9
2010-12-19s3: Fix some nonempty blank linesVolker Lendecke1-22/+22
2010-12-19s3: re-add some debugs to check_reduced_nameVolker Lendecke1-0/+2
2010-12-19s3: Fix some typosVolker Lendecke2-4/+4
2010-12-18Rename vfs operation posix_fallocate to just fallocate and add the ↵Jeremy Allison1-9/+12
vfs_fallocate_mode parameter. It turns out we need the fallocate operations to be able to both allocate and extend filesize, and to allocate and not extend filesize, and posix_fallocate can only do the former. So by defining the vfs op as posix_fallocate we lose the opportunity to use any underlying syscalls (like Linux fallocate) that can do the latter as well. We don't currently use the non-extending filesize call, but now I've changed the vfs op definition we can in the future. For the moment simply map the fallocate op onto posix_fallocate for the VFS_FALLOCATE_EXTEND_SIZE case and return ENOSYS for the VFS_FALLOCATE_KEEP_SIZE case. Jeremy. Autobuild-User: Jeremy Allison <jra@samba.org> Autobuild-Date: Sat Dec 18 08:59:27 CET 2010 on sn-devel-104
2010-12-17Move checks inside file_set_sparse() to allow it to be called from anywhere.Jeremy Allison2-19/+21
Autobuild-User: Jeremy Allison <jra@samba.org> Autobuild-Date: Fri Dec 17 21:56:38 CET 2010 on sn-devel-104
2010-12-17file_set_sparse needs to be a handle based call.Jeremy Allison2-11/+9
2010-12-17s3:smbd: implement FSCTL_SET_SPARSE more correctlyBjörn Jacke1-5/+43
this is a port of a patch from metze for 3.3: We don't do the "strict allocation" when the sparse bit isn't set, but that shouldn't matter. We now allow windows applications to set and unset the sparse bit. Note that in order to implement this 100% like described in [MS-FSA], we'd have to change our data model and support the sparse flag per stream.
2010-12-17s3:smbd: add file_set_sparse() functionBjörn Jacke1-0/+49
this is based on a patch for 3.3 from metze
2010-12-17s3: remove set_sparse_flagBjörn Jacke1-15/+1
we need to determine sparseness from the sparse flag we store not from the allocation size on the POSIX filesystem. This is how Windows works - in the first place sparseness is a file flag, not the allocation state of the file Signed-off-by: Stefan Metzmacher <metze@samba.org>
2010-12-17s3: add FILE_ATTRIBUTE_SPARSE to get_stat_dos_flagsBjörn Jacke1-0/+2
generated by cgit (git 2.34.1) at 2024-12-29 14:37:11 +0000