o Man page for dmsetup

man/dmsetup.8.in

@@ -0,0 +1,132 @@
+.TH DMSETUP 8 "Nov 29 2001" "Linux" "MAINTENTANCE COMMANDS"
+.SH NAME
+dmsetup \- low level logical volume management
+.SH SYNOPSIS
+.ad l
+.B dmsetup create
+.I <device name> <table path>
+.br
+.B dmsetup remove
+.I <device name>
+.br
+.B dmsetup suspend
+.I <device name>
+.br
+.B dmsetup resume
+.I <device name>
+.br
+.B dmsetup reload
+.I <device name> <table path>
+.br
+.B dmsetup info
+.I <device name>
+.ad b
+.SH DESCRIPTION
+dmsetup is used to manage mapped devices using the device-mapper
+driver.  Devices are created by loading a table that gives a target
+for ranges of sectors in the logical device.
+
+The first argument to dmsetup is a command, after this the mapped
+device name (not the full path).
+.SH Commands
+.RS
+.IP \fBcreate
+.I <device name> <table path>
+.br
+Attempts to create a device using the table file given.  If
+successful a device will appear as
+/dev/device-mapper/<device-name>.  See below for information
+on the table file format.
+.IP \fBremove
+.I <device name>
+.br
+Removes the device
+.IP \fBsuspend
+.I <device name>
+.br
+Suspends a device, there are two parts to this.  Firstly any
+io that has already been mapped by the device, but has not yet
+completed will be flushed.  Secondly any further io to that
+device will be postponed for as long as the device is suspended.
+.IP \fBresume
+.I <device name>
+.br
+Un-suspends a device, postponed io now get's re-queued for processing.
+.IP \fBreload
+.I <device name> <table path>
+.br
+This command will only work if a device is in the suspended state.
+It allows the user to change the mapping table for an existing device.
+.IP \fBinfo
+.I <device name>
+.br
+Output's some brief information about the device in the form:
+.br
+<SUSPENDED|ACTIVE>
+.br
+<open count>
+.br
+<major>,<minor>
+.br
+<target count>
+.SH Table format
+Each line of the table specifies a single target, and is of the form:
+.br
+<logical start sector> <num sectors> <target type> <target args...>
+.br
+At the moment there are 3 simple target types available (though your
+system may have more in the form of modules).
+
+.IP \fBlinear
+.I <destination device> <start sector>
+.br
+The traditional linear mapping.
+
+.IP \fBstriped
+.I <num stripes> <chunk size> [<destination #n> <start sector >]+
+.br
+Creates a striped area, 
+.br
+eg, striped 2 32 /dev/hda1 0 /dev/hdb1 0
+    will map the first chunk (16k) as follows:
+.br
+    LV chunk 1 -> hda1, chunk 1
+.br
+    LV chunk 2 -> hdb1, chunk 1
+.br
+    LV chunk 3 -> hda1, chunk 2
+.br
+    LV chunk 4 -> hdb1, chunk 2
+.br
+    etc.
+
+
+.IP \fBio-err
+.br
+Errors any io that goes to this area.  Useful for testing or
+creating devices with holes in them.
+
+
+.SH Examples
+
+
+# A table to join two disks together
+.br
+0 1028160 linear /dev/hda 0
+.br
+1028160 3903762 linear /dev/hdb 0
+
+
+
+# A table to stripe across the two disks, 
+.br
+# and add the spare space from
+.br
+# hdb to the back of the volume
+
+0 2056320 striped 2 32 /dev/hda 0 /dev/hdb 0
+.br
+2056320 2875602 linear /dev/hdb 1028160
+
+.SH AUTHORS
+Original version: Joe Thornber (thornber@sistina.com)