summaryrefslogtreecommitdiffstats
path: root/source3/smbd/server.c
Commit message (Collapse)AuthorAgeFilesLines
* s3:smbprofile: Replace sysv shmem with tdbVolker Lendecke2015-03-061-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | What? This patch gets rid of the central shared memory segment referenced by "profile_p". Instead, every smbd gets a static profile_area where it collects profiling data. Once a second, every smbd writes this profiling data into a record of its own in a "smbprofile.tdb". smbstatus -P does a tdb_traverse on this database and sums up what it finds. Why? At least in my perception sysv IPC has not the best reputation on earth. The code before this patch uses shmat(). Samba ages ago has developed a good abstraction of shared memory: It's called tdb. The main reason why I started this is that I have a request to become more flexible with profiling data. Samba should be able to collect data per share or per user, something which is almost impossible to do with a fixed structure. My idea is to for example install a profile area per share and every second marshall this into one tdb record indexed by share name. smbstatus -P would then also collect the data and either aggregate them or put them into individual per-share statistics. This flexibility in the data model is not really possible with one fixed structure. But isn't it slow? Well, I don't think so. I can't really prove it, but I do believe that on large boxes atomically incrementing a shared memory value for every SMB does show up due to NUMA effects. With this patch the hot code path is completely process-local. Once a second every smbd writes into a central tdb, this of course does atomic operations. But it's once a second, not on every SMB2 read. There's two places where I would like to improve things: With the current code all smbds wake up once a second. With 10,000 potentially idle smbds this will become noticable. That's why the current only starts the timer when something has changed. The second place is the tdb traverse: Right now traverse is blocking in the sense that when it has to switch hash chains it will block. With mutexes, this means a syscall. I have a traverse light in mind that works as follows: It assumes a locked hash chain and then walks the complete chain in one run without unlocking in between. This way the caller can do nonblocking locks in the first round and only do blocking locks in a second round. Also, a lot of syscall overhead will vanish. This way smbstatus -P will have almost zero impact on normal operations. Pair-Programmed-With: Stefan Metzmacher <metze@samba.org> Signed-off-by: Volker Lendecke <vl@samba.org> Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Ralph Boehme <slow@samba.org>
* s3:locking: add leases_db infrastructureVolker Lendecke2014-12-041-0/+5
| | | | | | | | | | | | | | Will enable us to solve the dynamic share path problem with leases on [homes]. We're also able to give the correct error codes when a lease key is re-used with a different file name. Pair-Programmed-With: Jeremy Allison <jra@samba.org> Signed-off-by: Volker Lendecke <vl@samba.org> Signed-off-by: Jeremy Allison <jra@samba.org> Reviewed-by: Stefan Metzmacher <metze@samba.org>
* s3:smbd: avoid a compiler warning in open_sockets_smbd()Stefan Metzmacher2014-11-251-1/+3
| | | | | Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
* s3:smbd: Use "smbd profiling level"Volker Lendecke2014-11-191-6/+6
| | | | | | | | Pair-Programmed-With: Stefan Metzmacher <metze@samba.org> Signed-off-by: Volker Lendecke <vl@samba.org> Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
* smbd: Simplify remove_child_pidVolker Lendecke2014-10-071-1/+1
| | | | | | | In this if-branch ret is always != 0. Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: David Disseldorp <ddiss@samba.org>
* profiling: Remove some #ifdefsVolker Lendecke2014-10-031-2/+0
| | | | | | | | | The DO_PROFILE_INC thingies already #define to nothing without WITH_PROFILE, and any sane compiler will just not compile the if-condition if there is no body to be executed. Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
* s3:smbd: use tevent_loop_wait() in the parent smbd process.Michael Adam2014-09-191-9/+41
| | | | | | | Pair-Programmed-With: Stefan Metzmacher <metze@samba.org> Signed-off-by: Michael Adam <obnox@samba.org> Signed-off-by: Stefan Metzmacher <metze@samba.org>
* smbd: Only DEBUG errors from messaging_cleanupVolker Lendecke2014-08-211-2/+5
| | | | | Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Ronnie sahlberg <ronniesahlberg@gmail.com>
* smbd: Use messaging_cleanup()Volker Lendecke2014-08-111-2/+2
| | | | | Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
* printing: only reload printer shares on client enumDavid Disseldorp2014-08-081-20/+0
| | | | | | | | | | | | | | | | | | | | | | | | | Currently, automatic printer share updates are handled in the following way: - Background printer process (BPP) forked on startup - Parent smbd and per-client children await MSG_PRINTER_PCAP messages - BPP periodically polls the printing backend for printcap data - printcap data written to printer_list.tdb - MSG_PRINTER_PCAP sent to all smbd processes following update - smbd processes all read the latest printer_list.tdb data, and update their share listings This procedure is not scalable, as all smbd processes hit printer_list.tdb in parallel, resulting in a large spike in CPU usage. This change sees smbd processes only update their printer share lists only when a client asks for this information, e.g. via NetShareEnum or EnumPrinters. Bug: https://bugzilla.samba.org/show_bug.cgi?id=10652 Suggested-by: Volker Lendecke <vl@samba.org> Signed-off-by: David Disseldorp <ddiss@samba.org> Reviewed-by: Andreas Schneider <asn@samba.org>
* lib: directory_create_or_exist() does not use "uid" parameterVolker Lendecke2014-07-281-1/+1
| | | | | Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
* lib/util: move memcache.[ch] to the toplevel 'samba-util' libraryStefan Metzmacher2014-07-181-1/+1
| | | | | | | | | | This is generic enough that it could be used in all code. Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Volker Lendecke <vl@samba.org> Autobuild-User(master): Stefan Metzmacher <metze@samba.org> Autobuild-Date(master): Fri Jul 18 15:43:33 CEST 2014 on sn-devel-104
* smbd: Make messaging_send_to_children staticChristof Schmitt2014-06-301-2/+2
| | | | | | | | | | This function is only used in server.c Signed-off-by: Christof Schmitt <cs@samba.org> Reviewed-by: Volker Lendecke <vl@samba.org> Autobuild-User(master): Volker Lendecke <vl@samba.org> Autobuild-Date(master): Mon Jun 30 17:20:00 CEST 2014 on sn-devel-104
* smbd: Use common function for FORCE_DIS and KILL_CLIENT_IP in parentChristof Schmitt2014-06-301-22/+11
| | | | | | | | | The parent smbd process only forwards the message to the child processes. Use a common function instead of two separate ones that do the same thing. Signed-off-by: Christof Schmitt <cs@samba.org> Reviewed-by: Volker Lendecke <vl@samba.org>
* messaging3: Make messaging_dgm_cleanup return 0/errnoVolker Lendecke2014-06-181-3/+3
| | | | | Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
* s3/s4: smbd, rpc, ldap, cldap, kdc services.Jeremy Allison2014-06-071-6/+10
| | | | | | | | | | | | | | Allow us to start if we bind to *either* :: or 0.0.0.0. Allows us to cope with systems configured as only IPv4 or only IPv6. Signed-off-by: Jeremy Allison <jra@samba.org> Reviewed-By: Amitay Isaacs <amitay@gmail.com> Reviewed-By: Alexander Bokovoy <ab@samba.org> Autobuild-User(master): Jeremy Allison <jra@samba.org> Autobuild-Date(master): Sat Jun 7 01:01:44 CEST 2014 on sn-devel-104
* s3: smbd: Call dgram cleanup init background setup.Jeremy Allison2014-04-231-0/+4
| | | | | Signed-off-by: Jeremy Allison <jra@samba.org> Reviewed-by: Michael Adam <obnox@samba.org>
* smbd: Always clean up the child's msg_ctxVolker Lendecke2014-04-231-6/+5
| | | | | | | | | | This is a bit lazy programming, we could and possibly should do this in exit_server() in the child. But this way we make sure the cleanup works. If it only was executed for unclean exits, we might not detect failure of this code in the parent. Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
* lib: Remove messages_localVolker Lendecke2014-04-231-11/+0
| | | | | Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
* lib: Add messaging_dgmVolker Lendecke2014-04-231-0/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | Messaging based on unix domain datagram sockets This makes every process participating in messaging bind on a unix domain datagram socket, similar to the source4 based messaging. The details are a bit different though: Retry after EWOULDBLOCK is done with a blocking thread, not by polling. This was the only way I could in experiments avoid a thundering herd or high load under Linux in extreme overload situations like many thousands of processes sending to one blocked process. If there are better ideas to do this in a simple way, I'm more than happy to remove the pthreadpool dependency again. There is only one socket per process, not per task. I don't think that per-task sockets are really necessary, we can do filtering in user space. The message contains the destination server_id, which contains the destination task_id. I think we can rebase the source4 based imessaging on top of this, allowing multiple imessaging contexts on top of one messaging_context. I had planned to do this conversion before this goes in, but Jeremy convinced me that this has value in itself :-) Per socket we also create a fcntl-based lockfile to allow race-free cleanup of orphaned sockets. This lockfile contains the unique_id, which in the future will make the server_id.tdb obsolete. Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
* smbd: use exit_daemon() to support reporting to systemd from smbdAlexander Bokovoy2014-04-231-25/+19
| | | | | | | BUG: https://bugzilla.samba.org/show_bug.cgi?id=10517 Signed-off-by: Alexander Bokovoy <ab@samba.org> Reviewed-by: Andreas Schneider <asn@samba.org>
* add systemd integrationAlexander Bokovoy2014-04-231-0/+4
| | | | | | | | | | | | | | | Add --with-systemd / --without-systemd options to check whether libsystemd-daemon library is available and use it to report service startup status to systemd for smbd/winbindd/nmbd and AD DC. The problem it solves is correct reporting of the Samba services at the point when they are ready to serve clients, important for high availability software integration. BUG: https://bugzilla.samba.org/show_bug.cgi?id=10517 Signed-off-by: Alexander Bokovoy <ab@samba.org> Reviewed-by: Andreas Schneider <asn@samba.org>
* s3: messages: Implement cleanup of dead records.Jeremy Allison2014-04-031-0/+7
| | | | | | | | | | | | When a smbd process dies, pending messages.tdb records for this process might not get cleaned up. Implement a cleanup for dead records that is triggered after a smbd dies uncleanly; the records for that PID are deleted. Based on a patchset from Christof Schmitt <cs@samba.org>. Signed-off-by: Jeremy Allison <jra@samba.org> Reviewed-by: Christof Schmitt <cs@samba.org>
* s3-smbd: Align debug level for the same information.Andreas Schneider2014-04-011-3/+5
| | | | | | | | | | BUG: https://bugzilla.samba.org/show_bug.cgi?id=10388 Signed-off-by: Andreas Schneider <asn@samba.org> Reviewed-by: Alexander Bokovoy <ab@samba.org> Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org> Autobuild-Date(master): Tue Apr 1 23:16:19 CEST 2014 on sn-devel-104
* s3:build: remove global CLUSTER_SUPPORT defineStefan Metzmacher2014-03-241-2/+0
| | | | | | | | | | All ctdb specific code is isolated in samba-cluster-support.so now. Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Michael Adam <obnox@samba.org> Autobuild-User(master): Michael Adam <obnox@samba.org> Autobuild-Date(master): Mon Mar 24 19:08:44 CET 2014 on sn-devel-104
* cmdline: Remove dynconfig hooks in command line processingAndrew Bartlett2014-02-121-1/+0
| | | | | | | | | | | | | | | | | | | | | | This removes the ability to set paths like the sbindir, bindir, and changes the tool for setting lockdir statedir etc to be via --option="lock dir=/var/lock". These were originally added by commit 90a6873b0570f2691ba8d8fd11154c856bdd4415 by James Peach <jpeach@samba.org> The important use case, qemu, does not use these options, but specifies these directories via an smb.conf. They are being removed to remove a layer from the loadparm system, now that options can be specified from the command line. It will also make it easier to generate the affected parameters from the XML documentation if this layer of indirection is removed. Andrew Bartlett Signed-off-by: Andrew Bartlett <abartlet@samba.org> Reviewed-by: Stefan Metzmacher <metze@samba.org> Autobuild-User(master): Stefan Metzmacher <metze@samba.org> Autobuild-Date(master): Wed Feb 12 16:42:14 CET 2014 on sn-devel-104
* param: rename lp function and variable from 'piddir' to 'pid_directory'Garming Sam2014-02-071-3/+3
| | | | | | Signed-off-by: Garming Sam <garming@catalyst.net.nz> Reviewed-by: Andrew Bartlett <abartlet@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
* param: rename lp function and variable from 'lockdir' to 'lock_directory'Garming Sam2014-02-071-2/+2
| | | | | | Signed-off-by: Garming Sam <garming@catalyst.net.nz> Reviewed-by: Andrew Bartlett <abartlet@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
* smbd: Make "num_children" available by smbcontrolVolker Lendecke2014-01-061-0/+16
| | | | | Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Christof Schmit <cs@samba.org>
* s3:smbd react on message that client should be disconnectedChristian Ambach2013-12-121-0/+13
| | | | | | | | | if MSG_SMB_KILL_CLIENT_IP message comes in and our client has the IP address given as argument, then shutdown the connection immediately Signed-off-by: Christian Ambach <ambi@samba.org> Reviewed-by: Christof Schmitt <cs@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
* s3-smbd_shim: Add become_authenticated_pipe_user().Andreas Schneider2013-10-291-0/+2
| | | | | Signed-off-by: Andreas Schneider <asn@samba.org> Reviewed-by: Stefan Metzmacher <metze@samba.org>
* s3:smbd: use new function cluster_probe_ok()Michael Adam2013-10-171-9/+3
| | | | | Signed-off-by: Michael Adam <obnox@samba.org> Reviewed-by: Volker Lendecke <vl@samba.org>
* smbd/winbindd: Do an early check if ctdbd is functionalVolker Lendecke2013-10-171-0/+11
| | | | | | | | | This will avoid panic calls when smbd and winbind is started in cluster mode before ctdb is functional. It still logs something sane at debug level 0, but it does not panic and core anymore. Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Michael Adam <obnox@samba.org>
* Fix bug #9166 - Starting smbd or nmbd with stdin from /dev/null results in ↵Jeremy Allison2013-06-201-1/+13
| | | | | | | | | | "EOF on stdin" Only install the stdin handler if it's a pipe or fifo. Signed-off-by: Jeremy Allison <jra@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
* s3:smbd: add a scavenger process for disconnected durable handlesGregor Beck2013-04-181-0/+5
| | | | | | | | Pair-Programmed-With: Stefan Metzmacher <metze@samba.org> Signed-off-by: Gregor Beck <gbeck@sernet.de> Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Michael Adam <obnox@samba.org>
* s3:smbd: s/struct timed_event/struct tevent_timerStefan Metzmacher2013-02-191-2/+2
| | | | | Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Michael Adam <obnox@samba.org>
* s3:smbd: s/struct event_context/struct tevent_contextStefan Metzmacher2013-02-191-2/+2
| | | | | Signed-off-by: Stefan Metzmacher <metze@samba.org> Reviewed-by: Michael Adam <obnox@samba.org>
* smbd: Fix a NULL vs false return errorVolker Lendecke2013-01-231-1/+1
| | | | Reviewed-by: Stefan Metzmacher <metze@samba.org>
* Use the new directory_create_or_exist_strict() function.Andreas Schneider2013-01-091-1/+1
| | | | Reviewed-by: Andrew Bartlett <abartlet@samba.org>
* smbd_open_one_socket does not use the messaging_context variable so why pass ↵Richard Sharpe2012-11-141-3/+0
| | | | | | | | | it in? Reviewed by: Jeremy Allison <jra@samba.org> Autobuild-User(master): Jeremy Allison <jra@samba.org> Autobuild-Date(master): Wed Nov 14 02:19:46 CET 2012 on sn-devel-104
* A small error message fix in source3/smbd/server.cRichard Sharpe2012-11-141-1/+1
| | | | | | | Removes some incorrect info from an error message (probably from its old place when it was copied). Reviewed by: Jeremy Allison <jra@samba.org>
* s3:smbd: don't call claim/yield_connection() in ↵Stefan Metzmacher2012-10-191-3/+0
| | | | | | | | | make_connection_snum/close_cnum() This was used to maintain the connections.tdb database which is being removed. We use info from the smbXsrv_tcon instead. Signed-off-by: Michael Adam <obnox@samba.org>
* s3:lib: remove unused sessionid_*() functionsGregor Beck2012-10-191-3/+0
| | | | | Signed-off-by: Stefan Metzmacher <metze@samba.org> Signed-off-by: Michael Adam <obnox@samba.org>
* s3:smbd: add exit_server to the smbd_shim hooksStefan Metzmacher2012-10-191-0/+3
| | | | | | | | | This is in preparation of moving sessionid_tdb and conn_tdb to smbd exclusively. metze Signed-off-by: Michael Adam <obnox@samba.org>
* s3:smbd: move initialization of the smbd_shim from smbd_init_globals() to main()Michael Adam2012-10-191-0/+15
| | | | | | This is in preparation of adding server exit hooks to the shim. Signed-off-by: Michael Adam <obnox@samba.org>
* s3fs-smbd: Make sure the registry is set up before we init printing.Andreas Schneider2012-10-041-4/+4
| | | | | Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org> Autobuild-Date(master): Thu Oct 4 12:06:29 CEST 2012 on sn-devel-104
* s3-smbd: Don't segfault if user specified ports out for range.Andreas Schneider2012-09-261-7/+9
| | | | | Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org> Autobuild-Date(master): Wed Sep 26 16:28:44 CEST 2012 on sn-devel-104
* s3:printing only do printing_subsystem_update when printing is enabledChristian Ambach2012-09-231-1/+4
| | | | | | | no point in calling this if printing was disabled and no spoolss service was started this hurts CTDB clusters as the smbds on the cluster nodes will fight for the single record in the TDB This fixes Bug 9197 - Disabling printing still makes smbd create and access printer_list.tdb
* param: Add startup checks for valid server role/binary combinationsAndrew Bartlett2012-08-231-0/+7
| | | | | | | This should eliminate confusion from our users about what they can expect to successfully run. Andrew Bartlett
* lib/param: Remove use of lp{cfg,}_socket_address outside the NBT client and ↵Andrew Bartlett2012-07-271-6/+3
| | | | | | | | | server In these other cases, control of the sockets to bind to can be obtained using "bind interfaces only = yes" and "interfaces = ". Andrew Bartlett
generated by cgit (git 2.34.1) at 2025-06-15 00:51:12 +0000