1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
|
<html><head><meta http-equiv="Content-Type" content="text/html; charset=ISO-8859-1"><title>Chapter 39. Samba Performance Tuning</title><link rel="stylesheet" href="samba.css" type="text/css"><meta name="generator" content="DocBook XSL Stylesheets V1.60.1"><link rel="home" href="samba-doc.html" title="SAMBA Project Documentation"><link rel="up" href="Appendixes.html" title="Part VI. Appendixes"><link rel="previous" href="Other-Clients.html" title="Chapter 38. Samba and other CIFS clients"><link rel="next" href="DNSDHCP.html" title="Chapter 40. DNS and DHCP Configuration Guide"></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 39. Samba Performance Tuning</th></tr><tr><td width="20%" align="left"><a accesskey="p" href="Other-Clients.html">Prev</a> </td><th width="60%" align="center">Part VI. Appendixes</th><td width="20%" align="right"> <a accesskey="n" href="DNSDHCP.html">Next</a></td></tr></table><hr></div><div class="chapter" lang="en"><div class="titlepage"><div><div><h2 class="title"><a name="speed"></a>Chapter 39. Samba Performance Tuning</h2></div><div><div class="author"><h3 class="author"><span class="firstname">Paul</span> <span class="surname">Cochrane</span></h3><div class="affiliation"><span class="orgname">Dundee Limb Fitting Centre<br></span><div class="address"><p><tt class="email"><<a href="mailto:paulc@dth.scot.nhs.uk">paulc@dth.scot.nhs.uk</a>></tt></p></div></div></div></div><div><div class="author"><h3 class="author"><span class="firstname">Jelmer</span> <span class="othername">R.</span> <span class="surname">Vernooij</span></h3><div class="affiliation"><span class="orgname">The Samba Team<br></span><div class="address"><p><tt class="email"><<a href="mailto:jelmer@samba.org">jelmer@samba.org</a>></tt></p></div></div></div></div><div><div class="author"><h3 class="author"><span class="firstname">John</span> <span class="othername">H.</span> <span class="surname">Terpstra</span></h3><div class="affiliation"><span class="orgname">Samba Team<br></span><div class="address"><p><tt class="email"><<a href="mailto:jht@samba.org">jht@samba.org</a>></tt></p></div></div></div></div></div><div></div></div><div class="toc"><p><b>Table of Contents</b></p><dl><dt><a href="speed.html#id2964209">Comparisons</a></dt><dt><a href="speed.html#id2964253">Socket options</a></dt><dt><a href="speed.html#id2964343">Read size</a></dt><dt><a href="speed.html#id2964393">Max xmit</a></dt><dt><a href="speed.html#id2964453">Log level</a></dt><dt><a href="speed.html#id2964483">Read raw</a></dt><dt><a href="speed.html#id2964567">Write raw</a></dt><dt><a href="speed.html#id2964630">Slow Logins</a></dt><dt><a href="speed.html#id2964659">Client tuning</a></dt><dt><a href="speed.html#id2964684">Samba performance problem due changing kernel</a></dt><dt><a href="speed.html#id2964716">Corrupt tdb Files</a></dt></dl></div><div class="sect1" lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="id2964209"></a>Comparisons</h2></div></div><div></div></div><p>
The Samba server uses TCP to talk to the client. Thus if you are
trying to see if it performs well you should really compare it to
programs that use the same protocol. The most readily available
programs for file transfer that use TCP are ftp or another TCP based
SMB server.
</p><p>
If you want to test against something like a NT or WfWg server then
you will have to disable all but TCP on either the client or
server. Otherwise you may well be using a totally different protocol
(such as NetBEUI) and comparisons may not be valid.
</p><p>
Generally you should find that Samba performs similarly to ftp at raw
transfer speed. It should perform quite a bit faster than NFS,
although this very much depends on your system.
</p><p>
Several people have done comparisons between Samba and Novell, NFS or
WinNT. In some cases Samba performed the best, in others the worst. I
suspect the biggest factor is not Samba vs some other system but the
hardware and drivers used on the various systems. Given similar
hardware Samba should certainly be competitive in speed with other
systems.
</p></div><div class="sect1" lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="id2964253"></a>Socket options</h2></div></div><div></div></div><p>
There are a number of socket options that can greatly affect the
performance of a TCP based server like Samba.
</p><p>
The socket options that Samba uses are settable both on the command
line with the <tt class="option">-O</tt> option, or in the <tt class="filename">smb.conf</tt> file.
</p><p>
The <a class="indexterm" name="id2964287"></a><i class="parameter"><tt>socket options</tt></i> section of the <tt class="filename">smb.conf</tt> manual page describes how
to set these and gives recommendations.
</p><p>
Getting the socket options right can make a big difference to your
performance, but getting them wrong can degrade it by just as
much. The correct settings are very dependent on your local network.
</p><p>
The socket option TCP_NODELAY is the one that seems to make the
biggest single difference for most networks. Many people report that
adding <a class="indexterm" name="id2964323"></a><i class="parameter"><tt>socket options</tt></i> = TCP_NODELAY doubles the read
performance of a Samba drive. The best explanation I have seen for this is
that the Microsoft TCP/IP stack is slow in sending tcp ACKs.
</p></div><div class="sect1" lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="id2964343"></a>Read size</h2></div></div><div></div></div><p>
The option <a class="indexterm" name="id2964353"></a><i class="parameter"><tt>read size</tt></i> affects the overlap of disk
reads/writes with network reads/writes. If the amount of data being
transferred in several of the SMB commands (currently SMBwrite, SMBwriteX and
SMBreadbraw) is larger than this value then the server begins writing
the data before it has received the whole packet from the network, or
in the case of SMBreadbraw, it begins writing to the network before
all the data has been read from disk.
</p><p>
This overlapping works best when the speeds of disk and network access
are similar, having very little effect when the speed of one is much
greater than the other.
</p><p>
The default value is 16384, but very little experimentation has been
done yet to determine the optimal value, and it is likely that the best
value will vary greatly between systems anyway. A value over 65536 is
pointless and will cause you to allocate memory unnecessarily.
</p></div><div class="sect1" lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="id2964393"></a>Max xmit</h2></div></div><div></div></div><p>
At startup the client and server negotiate a <i class="parameter"><tt>maximum transmit</tt></i> size,
which limits the size of nearly all SMB commands. You can set the
maximum size that Samba will negotiate using the <a class="indexterm" name="id2964414"></a><i class="parameter"><tt>max xmit</tt></i> option
in <tt class="filename">smb.conf</tt>. Note that this is the maximum size of SMB requests that
Samba will accept, but not the maximum size that the *client* will accept.
The client maximum receive size is sent to Samba by the client and Samba
honours this limit.
</p><p>
It defaults to 65536 bytes (the maximum), but it is possible that some
clients may perform better with a smaller transmit unit. Trying values
of less than 2048 is likely to cause severe problems.
</p><p>
In most cases the default is the best option.
</p></div><div class="sect1" lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="id2964453"></a>Log level</h2></div></div><div></div></div><p>
If you set the log level (also known as <a class="indexterm" name="id2964463"></a><i class="parameter"><tt>debug level</tt></i>) higher than 2
then you may suffer a large drop in performance. This is because the
server flushes the log file after each operation, which can be very
expensive.
</p></div><div class="sect1" lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="id2964483"></a>Read raw</h2></div></div><div></div></div><p>
The <a class="indexterm" name="id2964492"></a><i class="parameter"><tt>read raw</tt></i> operation is designed to be an optimised, low-latency
file read operation. A server may choose to not support it,
however. and Samba makes support for <a class="indexterm" name="id2964510"></a><i class="parameter"><tt>read raw</tt></i> optional, with it
being enabled by default.
</p><p>
In some cases clients don't handle <a class="indexterm" name="id2964528"></a><i class="parameter"><tt>read raw</tt></i> very well and actually
get lower performance using it than they get using the conventional
read operations.
</p><p>
So you might like to try <a class="indexterm" name="id2964549"></a><i class="parameter"><tt>read raw</tt></i> = no and see what happens on your
network. It might lower, raise or not affect your performance. Only
testing can really tell.
</p></div><div class="sect1" lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="id2964567"></a>Write raw</h2></div></div><div></div></div><p>
The <a class="indexterm" name="id2964577"></a><i class="parameter"><tt>write raw</tt></i> operation is designed to be an optimised, low-latency
file write operation. A server may choose to not support it,
however. and Samba makes support for <a class="indexterm" name="id2964594"></a><i class="parameter"><tt>write raw</tt></i> optional, with it
being enabled by default.
</p><p>
Some machines may find <a class="indexterm" name="id2964613"></a><i class="parameter"><tt>write raw</tt></i> slower than normal write, in which
case you may wish to change this option.
</p></div><div class="sect1" lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="id2964630"></a>Slow Logins</h2></div></div><div></div></div><p>
Slow logins are almost always due to the password checking time. Using
the lowest practical <a class="indexterm" name="id2964641"></a><i class="parameter"><tt>password level</tt></i> will improve things.
</p></div><div class="sect1" lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="id2964659"></a>Client tuning</h2></div></div><div></div></div><p>
Often a speed problem can be traced to the client. The client (for
example Windows for Workgroups) can often be tuned for better TCP
performance. Check the sections on the various clients in
<a href="Other-Clients.html" title="Chapter 38. Samba and other CIFS clients">Samba and Other Clients</a>.
</p></div><div class="sect1" lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="id2964684"></a>Samba performance problem due changing kernel</h2></div></div><div></div></div><p>
Hi everyone. I am running Gentoo on my server and samba 2.2.8a. Recently
I changed kernel version from linux-2.4.19-gentoo-r10 to
linux-2.4.20-wolk4.0s. And now I have performance issue with samba. Ok
many of you will probably say that move to vanilla sources...well I tried
it too and it didn't work. I have 100mb LAN and two computers (linux +
Windows2000). Linux server shares directory with DivX files, client
(windows2000) plays them via LAN. Before when I was running 2.4.19 kernel
everything was fine, but now movies freezes and stops...I tried moving
files between server and Windows and it's terribly slow.
</p><p>
Grab mii-tool and check the duplex settings on the NIC.
My guess is that it is a link layer issue, not an application
layer problem. Also run ifconfig and verify that the framing
error, collisions, etc... look normal for ethernet.
</p></div><div class="sect1" lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="id2964716"></a>Corrupt tdb Files</h2></div></div><div></div></div><p>
Well today it happened, Our first major problem using samba.
Our samba PDC server has been hosting 3 TB of data to our 500+ users
[Windows NT/XP] for the last 3 years using samba, no problem.
But today all shares went SLOW; very slow. Also the main smbd kept
spawning new processes so we had 1600+ running smbd's (normally we avg. 250).
It crashed the SUN E3500 cluster twice. After a lot of searching I
decided to <b class="command">rm /var/locks/*.tdb</b>. Happy again.
</p><p>
Q1) Is there any method of keeping the *.tdb files in top condition or
how to early detect corruption?
</p><p>
A1) Yes, run <b class="command">tdbbackup</b> each time after stopping nmbd and before starting nmbd.
</p><p>
Q2) What I also would like to mention is that the service latency seems
a lot lower then before the locks cleanup, any ideas on keeping it top notch?
</p><p>
A2) Yes! Same answer as for Q1!
</p></div></div><div class="navfooter"><hr><table width="100%" summary="Navigation footer"><tr><td width="40%" align="left"><a accesskey="p" href="Other-Clients.html">Prev</a> </td><td width="20%" align="center"><a accesskey="u" href="Appendixes.html">Up</a></td><td width="40%" align="right"> <a accesskey="n" href="DNSDHCP.html">Next</a></td></tr><tr><td width="40%" align="left" valign="top">Chapter 38. Samba and other CIFS clients </td><td width="20%" align="center"><a accesskey="h" href="samba-doc.html">Home</a></td><td width="40%" align="right" valign="top"> Chapter 40. DNS and DHCP Configuration Guide</td></tr></table></div></body></html>
|
