nixpkgs/doc/manual/troubleshooting.xml

<chapter xmlns="http://docbook.org/ns/docbook"
         xmlns:xlink="http://www.w3.org/1999/xlink">

<title>Troubleshooting</title>


<section>
  
<title>Debugging the boot process</title>

<para>To get a Stage 1 shell (i.e., a shell in the initial ramdisk),
add <literal>debug1</literal> to the kernel command line.  The shell
gets started before anything useful has been done.  That is, no
modules have been loaded and no file systems have been mounted, except
for <filename>/proc</filename> and <filename>/sys</filename>.</para>

<para>To get a Stage 2 shell (i.e., a shell in the actual root file
system), add <literal>debug2</literal> to the kernel command
line.  This shell is started right after stage 1 calls the stage 2
<literal>init</literal> script, so the root file system is there but
no services have been started.</para>

</section>



<section>
  
<title>Safe mode</title>

<para>If the hardware autodetection (in
<filename>upstart-jobs/hardware-scan</filename>) causes problems, add
<literal>safemode</literal> to the kernel command line.  This will
disable auto-loading of modules for your PCI devices.  However, you
will probably need to explicitly add modules to
<option>boot.kernelModules</option> to get network support etc.</para>

</section>


<section>
  
<title>Maintenance mode</title>

<para>You can go to maintenance mode by doing

<screen>
$ shutdown now</screen>

This will eventually give you a single-user root shell.

To get out of maintenance mode, do

<screen>
$ initctl emit startup</screen>

</para>

</section>



</chapter>