Fix glusterfs backend crash found at the Microsoft interop event.

[Samba.git] / docs-xml / Samba3-HOWTO / TOSHARG-Bugs.xml

<?xml version="1.0" encoding="iso-8859-1"?>
<!DOCTYPE chapter PUBLIC "-//Samba-Team//DTD DocBook V4.2-Based Variant V1.0//EN" "http://www.samba.org/samba/DTD/samba-doc">
<chapter id="bugreport">

<chapterinfo>
        &author.jht;
        &author.jelmer;
        &author.tridge;
        <pubdate> 27 June 1997 </pubdate>
</chapterinfo>

<title>Reporting Bugs</title>

<sect1>
<title>Introduction</title>

<para>
<indexterm><primary>Bugzilla</primary></indexterm>
<indexterm><primary>bug reports</primary></indexterm>
Please report bugs using Samba's <ulink url="https://bugzilla.samba.org/">Bugzilla</ulink> facilities and take
the time to read this file before you submit a bug report. Also, check to see if it has changed between
releases, as we may be changing the bug reporting mechanism at some point.
</para>

<para>
Please do as much as you can yourself to help track down the
bug. Samba is maintained by a dedicated group of people who volunteer
their time, skills, and efforts. We receive far more mail than
we can possibly answer, so you have a much higher chance of a response
and a fix if you send us a <quote>developer-friendly</quote> bug report that lets
us fix it fast. 
</para>

<para>
<indexterm><primary>configuration problem</primary></indexterm>
If you suspect that your 
problem is not a bug but a configuration problem, it is best to send 
it to the Samba mailing list, as there are thousands of other users on
that list who may be able to help you.
</para>

<para>
You may also like to look though the recent mailing list archives,
which are conveniently accessible on the Samba Web pages
at <ulink noescape="1" url="http://samba.org/samba/">http://samba.org/samba/</ulink>.
</para>

</sect1>

<sect1>
<title>General Information</title>

<para>
Before submitting a bug report, check your config for silly
errors. Look in your log files for obvious messages that tell
you've misconfigured something. Run testparm to check your config
file for correct syntax.
</para>

<para>
Have you looked through <link linkend="diagnosis">The Samba Checklist</link>? This is extremely important.
</para>

<para>
If you include part of a log file with your bug report, then be sure to
annotate it with exactly what you were doing on the client at the
time and exactly what the results were.
</para>

</sect1>

<sect1 id="dbglvl">
<title>Debug Levels</title>

<para>
If the bug has anything to do with Samba behaving incorrectly as a
server (like refusing to open a file), then the log files will probably
be quite useful. Depending on the problem, a log level of between 3 and
10 showing the problem may be appropriate. A higher level gives more
detail but may use too much disk space.
</para>

<para>
<indexterm><primary>debug level</primary></indexterm>
<indexterm><primary>log level</primary></indexterm>
To set the debug level, use the <smbconfoption name="log level"/> in your 
&smb.conf;. You may also find it useful to set the log 
level higher for just one machine and keep separate logs for each machine. 
To do this, add the following lines to your main &smb.conf; file:
</para>

<smbconfblock>
<smbconfoption name="log level">10</smbconfoption>
<smbconfoption name="log file">/usr/local/samba/lib/log.%m</smbconfoption>
<smbconfoption name="include">/usr/local/samba/lib/smb.conf.%m</smbconfoption>
</smbconfblock>

<para>
and create a file <filename>/usr/local/samba/lib/smb.conf.<replaceable>machine</replaceable></filename> where
<replaceable>machine</replaceable> is the name of the client you wish to debug. In that file put any
&smb.conf; commands you want; for example, <smbconfoption name="log level"/> may be useful. This also allows
you to experiment with different security systems, protocol levels, and so on, on just one machine.
</para>

<para>
The &smb.conf; entry <smbconfoption name="log level"/> is synonymous with the parameter <smbconfoption
name="debuglevel"/> that has been used in older versions of Samba and is being retained for backward
compatibility of &smb.conf; files.
</para>

<para>
As the <smbconfoption name="log level"/> value is increased, you will record a significantly greater level of
debugging information. For most debugging operations, you may not need a setting higher than
<constant>3</constant>. Nearly all bugs can be tracked at a setting of <constant>10</constant>, but be
prepared for a large volume of log data.
</para>

        <sect2>
        <title>Debugging-Specific Operations</title>

        <para>
<indexterm><primary>debugging</primary></indexterm>
<indexterm><primary>logging</primary></indexterm>
<indexterm><primary>functional components</primary></indexterm>
<indexterm><primary>cluttering</primary></indexterm>
        Samba-3.x permits debugging (logging) of specific functional components without unnecessarily
        cluttering the log files with detailed logs for all operations. An example configuration to 
        achieve this is shown in:
        </para>

<para>
<smbconfblock>
<smbconfoption name="log level">0 tdb:3 passdb:5 auth:4 vfs:2</smbconfoption>
<smbconfoption name="max log size">0</smbconfoption>
<smbconfoption name="log file">/var/log/samba/%U.%m.log</smbconfoption>
</smbconfblock>
</para>

        <para>
        This will cause the level of detail to be expanded to the debug class (log level) passed to
        each functional area per the value shown above. The first value passed to the <parameter>log level</parameter>
        of <constant>0</constant> means turn off all unnecessary debugging except the debug classes set for
        the functional areas as specified. The table shown in <link linkend="dbgclass">Debuggable Functions</link>
        may be used to attain very precise analysis of each SMB operation Samba is conducting.
        </para>

        <table frame="all" id="dbgclass">
                <title>Debuggable Functions</title>
        <tgroup cols="2" align="center">
                <thead>
                <row><entry>Function Name</entry><entry>Function Name</entry></row>
                </thead>
                <tbody>
                <row><entry>all</entry><entry>passdb</entry></row>
                <row><entry>tdb</entry><entry>sam</entry></row>
                <row><entry>printdrivers</entry><entry>auth</entry></row>
                <row><entry>lanman</entry><entry>winbind</entry></row>
                <row><entry>smb</entry><entry>vfs</entry></row>
                <row><entry>rpc_parse</entry><entry>idmap</entry></row>
                <row><entry>rpc_srv</entry><entry>quota</entry></row>
                <row><entry>rpc_cli</entry><entry>acls</entry></row>
                </tbody>
        </tgroup>
        </table>

        </sect2>

</sect1>

<sect1>
<title>Internal Errors</title>

<para>
If you get the message <quote><errorname>INTERNAL ERROR</errorname></quote> in your log files, 
it means that Samba got an unexpected signal while running. It is probably a
segmentation fault and almost certainly means a bug in Samba (unless
you have faulty hardware or system software).
</para>

<para>
If the message came from smbd, it will probably be accompanied by
a message that details the last SMB message received by smbd. This
information is often useful in tracking down the problem, so please
include it in your bug report.
</para>

<para>
You should also detail how to reproduce the problem, if
possible. Please make this reasonably detailed.
</para>


<para>
<indexterm><primary>core files</primary></indexterm>
You may also find that a core file appeared in a <filename>corefiles</filename>
subdirectory of the directory where you keep your Samba log
files. This file is the most useful tool for tracking down the bug. To
use it, you do this:
<indexterm><primary>gdb</primary></indexterm>
<indexterm><primary>debug</primary></indexterm>
<screen>
&prompt;<userinput>gdb smbd core</userinput>
</screen>
</para>

<para>
<indexterm><primary>dbx</primary></indexterm>
<indexterm><primary>stack trace</primary></indexterm>
adding appropriate paths to smbd and core so gdb can find them. If you
do not have gdb, try <userinput>dbx</userinput>. Then within the debugger,
use the command <command>where</command> to give a stack trace of where the
problem occurred. Include this in your report.
</para>

<para>
<indexterm><primary>disass</primary></indexterm>
If you know any assembly language, do a <command>disass</command> of the routine
where the problem occurred (if it's in a library routine, then
disassemble the routine that called it) and try to work out exactly
where the problem is by looking at the surrounding code. Even if you
do not know assembly, including this information in the bug report can be
useful. 
</para>
</sect1>

<sect1>
<title>Attaching to a Running Process</title>

<para>
<indexterm><primary>PID</primary></indexterm>
<indexterm><primary>gdb</primary></indexterm>
<indexterm><primary>smbstatus</primary></indexterm>
Unfortunately, some UNIXes (in particular some recent Linux kernels)
refuse to dump a core file if the task has changed UID (which smbd
does often). To debug with this sort of system, you could try to attach
to the running process using
<userinput>gdb smbd <replaceable>PID</replaceable></userinput>, where you get
<replaceable>PID</replaceable> from <application>smbstatus</application>.
Then use <command>c</command> to continue and try to cause the core dump
using the client. The debugger should catch the fault and tell you
where it occurred.
</para>

<para>
Sometimes it is necessary to build Samba binary files that have debugging
symbols so as to make it possible to capture enough information from a crashed
operation to permit the Samba Team to fix the problem.
</para>

<para>
Compile with <constant>-g</constant> to ensure you have symbols in place. 
Add the following line to the &smb.conf; file global section:
<screen>
panic action = "/bin/sleep 90000"
</screen>
to catch any panics. If <command>smbd</command> seems to be frozen, look for any sleep
processes. If it is not, and appears to be spinning, find the PID
of the spinning process and type:
<screen>
&rootprompt; gdb -p PID
</screen>
<indexterm><primary>spinning process</primary></indexterm>
then type <quote>bt full</quote> to
get a backtrace to see where the smbd is in the call path.
</para>

</sect1>

<sect1>
<title>Patches</title>


<para>
<indexterm><primary>diff</primary></indexterm>
<indexterm><primary>patch</primary></indexterm>
The best sort of bug report is one that includes a fix! If you send us
patches, please use <userinput>diff -u</userinput> format if your version of 
diff supports it; otherwise, use <userinput>diff -c4</userinput>. Make sure 
you do the diff against a clean version of the source and let me know 
exactly what version you used. 
</para>

</sect1>
</chapter>