src/nash/nash.8

   1 .TH NASH 8 "Mon Aug 02 2004"
   2 .SH NAME
   3 nash \- script interpretor to interpret linuxrc images
   4 .SH SYNOPSIS
   5 \fBnash\fR [--quiet] [--force] \fIscript\fR
   6
   7 .SH DESCRIPTION
   8 \fBnash\fR is a very simple script interpretor designed to be as small as
   9 possible. It is primarily designed to run simple linuxrc scripts on an initrd
  10 image. Arguments to commands may be enclosed in either single or double
  11 quotes to allow spaces to be included in the arguments. Spaces outside of
  12 quotations always delineate arguments, and so backslash escaping is supported.
  13
  14 Additionally, if nash is invoked as \fBmodprobe\fR, it will immediately
  15 exit with a return code of zero. This is to allow initrd's to prevent some
  16 extraneous kernel error messages during startup.
  17
  18 There are two types of commands, built in and external. External commands
  19 are run from the filesystem via execve(). If commands names are given without
  20 a path, nash will search it's builtin PATH, which is /usr/bin, /bin,
  21 /sbin, /usr/sbin.
  22
  23 Currently, \fBnash\fR supports the following built in commands.
  24
  25 .TP
  26 \fBaccess -[r][w][x][f] \fIpath\fR
  27 Tells whether the current user has sufficient permissions to read, write, or
  28 execture \fIpath\fR, or if the file exists (see \fBaccess(2)\fR for more
  29 information).
  30
  31 .TP
  32 \fBecho [item]* [> \fIfilename\fR]\fR
  33 Echos the text strings given to a file, with a space in between each
  34 item. The output may be optionally redirected to a file.
  35
  36 \fBexec\fR \fI<command>\fR
  37 The command given is execed, overlaying the nash process.
  38
  39 .TP
  40 \fBfind \fIdir\fR -name \fIname\fR
  41 Display the path to files named \fIname\fR in or below directory \fIdir\FR.
  42 This is a \fBvery\fR limited implementation of find(1).
  43
  44 .TP
  45 \fBfindlodev\fR
  46 Prints the full path to the first unused loopback block device on the
  47 system. If none is available, no output is displayed.
  48
  49 .TP
  50 \fBlosetup \fI/dev/loopdev\fR \fIfile\fR
  51 Binds \fIfile\fR to the loopback device \fI/dev/loopdev\fR. See
  52 \fBlosetup(8)\fR for information on loopback devices.
  53
  54 .TP
  55 \fBmkdevices\fR \fIpath\fR
  56 Creates device files for all of the block devices listed in
  57 \fB/proc/partitions\fR in the directory specfied by \fIpath\fR.
  58
  59 .TP
  60 \fBmkdir\fR \fI[-p]\fR \fIpath\fR
  61 Creates the directory \fIpath\fR. If \fI-p\fR is specified, this command
  62 will not complain if the directory exists. Note this is a subset of the
  63 standard \fBmkdir\fR \fI-p\fR behavior.
  64
  65 .TP
  66 \fBmknod\fR \fIpath\fR \fI[c|b]\fR \fImajor\fR \fIminor\fR
  67 Creates a device inode for \fIpath\fR. This is identical to \fBmkdev\fR(1)
  68 which the exceptions that it will not create named pipes and if the directories
  69 in \fIpath\fR do not exist they will be automatically created.
  70
  71 .TP
  72 \fBmkdmnod\fR
  73 Creates a device inode for the device mapper control inode as
  74 \fI/dev/mapper/control\fR.  If it already exists with the correct
  75 major/minor, it will not be recreated.
  76
  77 .TP
  78 \fBmkrootdev \fIpath\fR
  79 Makes \fIpath\fR a block inode for the device which should be mounted
  80 as root. To determine this device nash uses
  81 the device suggested by the root= kernel command line argument (if
  82 root=LABEL is used devices are probed to find one with that label). If
  83 no root= argument is available, /proc/sys/kernel/real-root-dev provides
  84 the device number.
  85
  86 .TP
  87 \fBmount \fI[--ro]\fR -o \fIopts\fR -t \fItype\fR \fIdevice\fR \fImntpoint\fR
  88 Mounts a filesystem. It does not support NFS, and it must be used in
  89 the form given above (arguments must go first).  If \fIdevice\fR is of the form
  90 \fBLABEL=\fIfoo\fR the devices listed in /fB/proc/partitions\fR will
  91 be searched, and the first device with a volume label of \fIfoo\fR will
  92 be mounted. Normal \fBmount\fR(2) options are supported, and \fB--ro\fR will
  93 mount the filesystem read only for compatibility with older versions of nash.
  94 The \fBdefaults\fR mount option is silently ignored.
  95
  96 .TP
  97 \fBpivot_root \fInewrootpath\fR \fIoldrootpath\fR
  98 Makes the filesystem mounted at \fInewrootpath\fR the new root filesystem,
  99 and mounts the current root filesystem as \fIoldrootpath\fR.
 100
 101 .TP
 102 \fBreadlink \fIpath\fR
 103 Displays the value of the symbolic link \fIpath\fR.
 104
 105 .TP
 106 \fBraidautorun \fImddevice\fR
 107 Runs raid autodetection on all raid-typed partitions. \fImddevice\fR must
 108 be a raid device (any will do).
 109
 110 .TP
 111 \fBsetquiet\fR
 112 Cause any later echos in this script to not be displayed.
 113
 114 .TP
 115 \fBshowlabels\fR
 116 Display a table of devices, their filesystem labels, and their uuids.
 117
 118 .TP
 119 \fBsleep \fInum\fR
 120 Sleep for \fInum\fR seconds
 121
 122 .TP
 123 \fBswitchroot \fInewrootpath\fR
 124 Makes the filesystem mounted at \fInewrootpath\fR the new root
 125 filesystem by moving the mountpoint.  This will only work in 2.6 or
 126 later kernels.
 127
 128 .TP
 129 \fBumount \fIpath\fR
 130 Unmounts the filesystem mounted at \fIpath\fR.
 131
 132 .SH RETURN VALUE
 133 Returns 0 is the last command succeeded or 1 if it failed.
 134
 135 .SH OPTIONS
 136 .TP
 137 \fB-\-force\fR
 138 Allows \fBforce\fR really execute the script, even though \fBnash\fR doesn't
 139 appear to be running from an initrd image.
 140
 141 .SH BUGS
 142 Probably many. \fBnash\fR is \fbnot a shell\fR, and it shouldn't be thought of
 143 as one. It isn't entirely different from a shell, but that's mostly by
 144 accident.
 145
 146 .SH AUTHOR
 147 .nf
 148 Erik Troan <ewt@redhat.com>
 149 .fi