Go back --> [wiki:Tutorial Tutorials]
[[TOC(heading=Tutorial TOC, Tutorial, Tutorial/HowToSave, depth=2)]]

= How to save an existing disk image from one node of a Testbed =

If you have not done so yet: 

 * [http://www.orbit-lab.org/wiki/registration Register] for an account
 * [https://www.orbit-lab.org/schedule/ Make a reservation] on the Scheduler for a given testbed

Lets assume that you registered as user 'bob' and made a reservation for the 'sb1' testbed. You then access the 'sb1' console using the command:

 {{{
  ssh bob@console.sb1.orbit-lab.org
 }}}
[[BR]]
== Getting Started ==

Then, to save the existing disk image on node (1,1) of the 'sb1' testbed, use the command:

 {{{
  orbit save [1,1] 
  # will save the current disk image on node [1,1] of the 'sb1' testbed

 }}}

The output of this image saving process will look like the following:

 {{{

Domain information from the console: 'sb1'
Saving image from node '[1,1]' from domain 'sb1'... INFO init: NodeHandler Version 4.2.0 (1272)
 INFO init: Experiment ID: sb1_2008_07_20_23_38_04
 ...
 INFO stdlib: Waiting for nodes (Up/Down/Total): 0/1/1 - (still down: n_1_1)
 ...
 INFO n_1_1: - Saving disk image from node n_1_1 in the file 'node-1-1-2008-07-20-23-39-39.ndz'
 INFO n_1_1:   (disk images are located at: 'frisbee:/export/orbit/image/tmp')
 INFO prop.started: started = "true":String
 INFO exp: - SAVE_IMAGE process started at: Sun Jul 20 23:39:49 -0400 2008
 INFO exp:   (this may take a while, e.g. 5min+, depending of the size of your image)
 INFO exp: - Saving process finished correctly at: Sun Jul 20 23:47:20 -0400 2008
 INFO Experiment: DONE!
 INFO ExecApp: Application 'commServer' finished
 INFO run: Experiment sb1_2008_07_20_23_38_04 finished after 9:19
 done.

 }}}

At the end of the ''saving'' process, you will have disk image file with the name: ''"node-1-1-2008-07-20-23-39-39.ndz"'' in the directory ''"/export/orbit/image/tmp"'' on the machine with the host name ''"frisbee"''. These information are all provided in the output displayed above.

You can then:
 * install this disk image on a set of nodes, using the instructions described in this [wiki:Tutorial/HowToImage tutorial]. In this example, to install the newly created disk image on node (1,2):
{{{
  orbit load [1,2] tmp/node-1-1-2008-07-20-23-39-39.ndz
  # will install the disk image on node [1,2] of the 'sb1' testbed

 }}}
 * only for the purpose of making a backup of your image: log into the ''"frisbee"'' machine, and copy your image to your backup storage.

[[BR]]
== Learning More ==

The above disk image saving process is implemented as a ''special'' orbit experiment. As such, its execution will results in a log file as with any other orbit experiment. This log file should look like  [attachment:sb1_2008_07_20_23_38_04.log this]. 

Each image created by the above saving process is a full hard-disk image, which can have arbitrary large file size (>200Mo or even more). As storage on the ''"frisbee"'' machine is limited, please be considerate in the number of images you save/use, and move any unused images to your own archival storage.

The generic '''orbit''' command used above is the access point to control various ORBIT functions, such as the saving of a disk image from a node, with the sub-command "save". To see a list of all the available '''orbit''' commands, you should type ''"orbit help"''.

Finally, the complete available option/commands for the ''save'' function are given by ''"orbit help save"'':

 {{{
  orbit help save


Save a disk image from a given node into an archive file

Usage:
      orbit save [-h]
      orbit save [NODE] [DOMAIN]
 
      With: 
      -h, --help    print this help message
      NODE          a valid node description 
                    (no default here, you have to enter a node!)
 
      Some Examples: 
                    orbit save [5,10]
                    orbit save [11,2] grid

 }}}