source: trunk/samba/docs/htmldocs/Samba3-Developers-Guide/architecture.html @ 30

Last change on this file since 30 was 1, checked in by Paul Smedley, 14 years ago

Initial code import

File size: 8.6 KB
1<html><head><meta http-equiv="Content-Type" content="text/html; charset=ISO-8859-1"><title>Chapter 3. Samba Architecture</title><link rel="stylesheet" href="samba.css" type="text/css"><meta name="generator" content="DocBook XSL Stylesheets V1.68.1"><link rel="start" href="index.html" title="SAMBA Developers Guide"><link rel="up" href="pt02.html" title="Part II. Samba Basics"><link rel="prev" href="pt02.html" title="Part II. Samba Basics"><link rel="next" href="debug.html" title="Chapter 4. The samba DEBUG system"></head><body bgcolor="white" text="black" link="#0000FF" vlink="#840084" alink="#0000FF"><div class="navheader"><table width="100%" summary="Navigation header"><tr><th colspan="3" align="center">Chapter 3. Samba Architecture</th></tr><tr><td width="20%" align="left"><a accesskey="p" href="pt02.html">Prev</a> </td><th width="60%" align="center">Part II. Samba Basics</th><td width="20%" align="right"> <a accesskey="n" href="debug.html">Next</a></td></tr></table><hr></div><div class="chapter" lang="en"><div class="titlepage"><div><div><h2 class="title"><a name="architecture"></a>Chapter 3. Samba Architecture</h2></div><div><div class="author"><h3 class="author"><span class="firstname">Dan</span> <span class="surname">Shearer</span></h3></div></div><div><p class="pubdate"> November 1997</p></div></div></div><div class="toc"><p><b>Table of Contents</b></p><dl><dt><span class="sect1"><a href="architecture.html#id292357">Introduction</a></span></dt><dt><span class="sect1"><a href="architecture.html#id292396">Multithreading and Samba</a></span></dt><dt><span class="sect1"><a href="architecture.html#id292420">Threading smbd</a></span></dt><dt><span class="sect1"><a href="architecture.html#id292473">Threading nmbd</a></span></dt><dt><span class="sect1"><a href="architecture.html#id292512">nbmd Design</a></span></dt></dl></div><div class="sect1" lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="id292357"></a>Introduction</h2></div></div></div><p>
2This document gives a general overview of how Samba works
3internally. The Samba Team has tried to come up with a model which is
4the best possible compromise between elegance, portability, security
5and the constraints imposed by the very messy SMB and CIFS
8It also tries to answer some of the frequently asked questions such as:
9</p><div class="orderedlist"><ol type="1"><li><p>
10        Is Samba secure when running on Unix? The xyz platform?
11        What about the root priveliges issue?
12</p></li><li><p>Pros and cons of multithreading in various parts of Samba</p></li><li><p>Why not have a separate process for name resolution, WINS, and browsing?</p></li></ol></div></div><div class="sect1" lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="id292396"></a>Multithreading and Samba</h2></div></div></div><p>
13People sometimes tout threads as a uniformly good thing. They are very
14nice in their place but are quite inappropriate for smbd. nmbd is
15another matter, and multi-threading it would be very nice.
17The short version is that smbd is not multithreaded, and alternative
18servers that take this approach under Unix (such as Syntax, at the
19time of writing) suffer tremendous performance penalties and are less
20robust. nmbd is not threaded either, but this is because it is not
21possible to do it while keeping code consistent and portable across 35
22or more platforms. (This drawback also applies to threading smbd.)
24The longer versions is that there are very good reasons for not making
25smbd multi-threaded.  Multi-threading would actually make Samba much
26slower, less scalable, less portable and much less robust. The fact
27that we use a separate process for each connection is one of Samba's
28biggest advantages.
29</p></div><div class="sect1" lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="id292420"></a>Threading smbd</h2></div></div></div><p>
30A few problems that would arise from a threaded smbd are:
31</p><div class="orderedlist"><ol type="1"><li><p>
32        It's not only to create threads instead of processes, but you
33        must care about all variables if they have to be thread specific
34        (currently they would be global).
36        if one thread dies (eg. a seg fault) then all threads die. We can
37        immediately throw robustness out the window.
39        many of the system calls we make are blocking. Non-blocking
40        equivalents of many calls are either not available or are awkward (and
41        slow) to use. So while we block in one thread all clients are
42        waiting. Imagine if one share is a slow NFS filesystem and the others
43        are fast, we will end up slowing all clients to the speed of NFS.
45        you can't run as a different uid in different threads. This means
46        we would have to switch uid/gid on _every_ SMB packet. It would be
47        horrendously slow.
49        the per process file descriptor limit would mean that we could only
50        support a limited number of clients.
52        we couldn't use the system locking calls as the locking context of
53        fcntl() is a process, not a thread.
54</p></li></ol></div></div><div class="sect1" lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="id292473"></a>Threading nmbd</h2></div></div></div><p>
55This would be ideal, but gets sunk by portability requirements.
57Andrew tried to write a test threads library for nmbd that used only
58ansi-C constructs (using setjmp and longjmp). Unfortunately some OSes
59defeat this by restricting longjmp to calling addresses that are
60shallower than the current address on the stack (apparently AIX does
61this). This makes a truly portable threads library impossible. So to
62support all our current platforms we would have to code nmbd both with
63and without threads, and as the real aim of threads is to make the
64code clearer we would not have gained anything. (it is a myth that
65threads make things faster. threading is like recursion, it can make
66things clear but the same thing can always be done faster by some
67other method)
69Chris tried to spec out a general design that would abstract threading
70vs separate processes (vs other methods?) and make them accessible
71through some general API. This doesn't work because of the data
72sharing requirements of the protocol (packets in the future depending
73on packets now, etc.) At least, the code would work but would be very
74clumsy, and besides the fork() type model would never work on Unix. (Is there an OS that it would work on, for nmbd?)
76A fork() is cheap, but not nearly cheap enough to do on every UDP
77packet that arrives. Having a pool of processes is possible but is
78nasty to program cleanly due to the enormous amount of shared data (in
79complex structures) between the processes. We can't rely on each
80platform having a shared memory system.
81</p></div><div class="sect1" lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="id292512"></a>nbmd Design</h2></div></div></div><p>
82Originally Andrew used recursion to simulate a multi-threaded
83environment, which use the stack enormously and made for really
84confusing debugging sessions. Luke Leighton rewrote it to use a
85queuing system that keeps state information on each packet.  The
86first version used a single structure which was used by all the
87pending states.  As the initialisation of this structure was
88done by adding arguments, as the functionality developed, it got
89pretty messy.  So, it was replaced with a higher-order function
90and a pointer to a user-defined memory block.  This suddenly
91made things much simpler: large numbers of functions could be
92made static, and modularised.  This is the same principle as used
93in NT's kernel, and achieves the same effect as threads, but in
94a single process.
96Then Jeremy rewrote nmbd. The packet data in nmbd isn't what's on the
97wire. It's a nice format that is very amenable to processing but still
98keeps the idea of a distinct packet. See "struct packet_struct" in
99nameserv.h.  It has all the detail but none of the on-the-wire
100mess. This makes it ideal for using in disk or memory-based databases
101for browsing and WINS support.
102</p></div></div><div class="navfooter"><hr><table width="100%" summary="Navigation footer"><tr><td width="40%" align="left"><a accesskey="p" href="pt02.html">Prev</a> </td><td width="20%" align="center"><a accesskey="u" href="pt02.html">Up</a></td><td width="40%" align="right"> <a accesskey="n" href="debug.html">Next</a></td></tr><tr><td width="40%" align="left" valign="top">Part II. Samba Basics </td><td width="20%" align="center"><a accesskey="h" href="index.html">Home</a></td><td width="40%" align="right" valign="top"> Chapter 4. The samba DEBUG system</td></tr></table></div></body></html>
Note: See TracBrowser for help on using the repository browser.