Converting a physical Blade HPUX OS to a HPUX VM

[Matthew Murdock]

I know that ignite/ux can be used to clone a exisiting hpux server OS to be used to create a new HPUX VM. However, I dont have the knowledge (mostly forgot all of it),  for all the intermediate steps to make it happen. 

I know you run ignite against the physcial box to generate a image (from the HPVM Server). At this point I need to have a re-edufication on using that image to be used as a deployable image (been way to long to remember). 

I do not know the full steps how to get that image to be used as a install depot, or pushed to a VM disk that can be attached to the VM as a boot disk. 

My HPVM Server is also a local IGNITE REPO server that serves out the HPUX OS to install from and it works. I have installed vm's from it. really the only big issue is with all the GNU packages installed on the client that I dont want to deal with. the Data can be scp'd across the network as needed.

I do have a HPUX support contract to work against, but, thought I could start here first to see how far I can get without tying up the Support Queues. 

Physical Client - BL860C HPUX Blade Gen1 running HPUX 11.31

Physical VM/Ignite Server - RX2800 I4 2x8core. running hpux 11.31 with  6.50HPVM.

[shiva_jr]

Hi @Matthew Murdock,

If you want to take the image by Ignite scripts then you can use the scrips 'make_sys_image' and 'save_config' to create image.  You can get the man page.
* Use the man page for “make_sys_tape” or “make_medeilif” for creating image using tape drive.
* Use the man page for “make_net_recovery” for creating image using network drive.
   Ex. : make_net_recovery -p -a <IP of server location to save archive>:<archive path> -s <IgniteUX Server>

Here are some of the posts and videos to help you.
1) Creating a Golden image using HPE Ignite-UX server
2) How to create the Ignite-UX client system recovery archive by using make_net_recovery
3) Booting HP-UX system using the hpe ignite-ux server and restoring the OS

Regards,
Shiva_JR
Please mark as 'Accepted solution' and thumbs up with Kudos if my post worked

[Matthew Murdock]

@shiva_jr 

The problem i am running into is the the size of  the client /var or the server /var  has to be increased to the accomodate the image (roughly 50GB used). I have created a NFS share of 100gb then I cannot mount the /var/opt/ignite/recovery/archives on the nfs share on the server (archives busy). 

With make_net_recovery, i cannot export the NFS share in /etc/dfs/dfstab.

srv:>shareall
share_nfs: Cannot share remote filesystem: /var/opt/ignite/recovery/archive

Otherwise, vi /etc/exports. The /etc/exports file on "srv" should contain the entry: "/var/opt/ignite/recovery/archive
-anon=2,access=client".  <== the archive directory is the nfs mount point mounted on srv.

I can mount the nfs point "archive" on the client to run make_net_recovery, but do not have the syntax to do that

make_net_recovery -s srv -a client:/var/opt/ignite/recovery/archive -x inc_entire=vg00     .<== Maybe??

[georgek_1]

Hello Matthew,

You are right as you could mention a different archive server / NFS share while performing make_net_recovery .
-s will point to the ignite server where you will have the ignite configuration , at /var/opt/ignite/clients/<clinet_name>
-a will save the archive to this directory when mentioned with make_net_recovery .
if -a is not mentioned , the archive will be saved to default location which is /var/opt/ignite/recovery/archives/<clinet_name>

//from man make_net_recovery
           -a archive_server:archive_directories
                Specifies the NFS server and location onto which to store
                the archive. The archive directory must be NFS exported (see
                the section Exporting Archive Directory), and sufficient
                disk space is required.  The default is the hostname of the
                Ignite-UX server followed by the directory which holds the
                archive, e.g.,
                Serverhost:/var/opt/ignite/recovery/archives/hostname.  The
                hostname is the name of the system being archived.  Each
                make_net_recovery client will create a subdirectory named
                for the client hostname under the specified directory to
                store the archives.

For example , You may take Ignite backup with following command , where
myserver = Ignite server , where you have the client configuration saved
nfs_server = Your NFS server where an NFS share (/archive) available for this client .

make_net_recovery -s myserver -a nfs_server:/archive -x inc_entire=vg00 

I work for HPE/ I am an HPE Employee (HPE Community)

[Matthew Murdock]

Thank you. now as far as space is concerned, if 200gb of vg00 is allocated, but only 50gb is used, is the backup 200gb or only 50gb. Personally, i think space to accomodate 200gb is required for the archive location?

[georgek_1]

Hello Matthew,

If  usage of vg00 is 50gb , then 50gb +/- 10% free space in archive location  should be enough to create the backup . Ignite using tar/pax to create  archive of actual data , hence 50+/-10% should be good .

I work for HPE/ I am an HPE Employee (HPE Community)