Setting up a backup machine

[Peter Clarke]

Hi,

I have an ES40 that we have been using for years now and have just purchased a new ES40 to use as a backup machine.What i need to do is get the new machine running exactly the same as the original machine with all the same software etc... we don't really need all the data files.
What would be the best way to achieve this?

Thanks

[Volker Halle]

Peter,

just take a copy of the system disk (and the otehr disks eventually) from your old ES40 and copy it to your new ES40. When you start up the new ES40, make sure you are not connected to the network to prevent network address conflicts.

Volker.

[Hein van den Heuvel]

Just clone the drives (BACKUP/IMAGE or BACKUP/PHYSICAL) shoudl do.

However... I'd urge you to take this opportunity to excercise the catastrophic recovery playbook.

What is the current procedure if the current box were to ever become unuseable?
Use that procedure on the target backup box!

Roll out those tapes!

Cheers,
Hein.

[Peter Clarke]

Would this take care of TCPIP, DECNET etc... or would i have to install these seperately?

[Jan van den Ende]

Peter,

Anything installed on the source system, will be at the copy.

However, ALL configs are also identical!

That might be your intention, but then you will have to make D**N sure that those systems will ***NEVER*** be able to "see" each other on whatever circuit!!
You can also temporarily isolate them, and then configure nodename, network adresses, disknames, custer ID, etc somewhat differently.
After that, you can setup things to, eg, be reachable from the same terminal (or emulator).

Also, when the copied system is on line, you will want to put its own licenses in its LMM database. :-)

hth

Proost.

Have one on me.

jpe

[Hein van den Heuvel]

Yeah, if your backups are correctly made, and the restores properly done, then everything should come back.

And that's exactly what the earlier warnign was about... you probably do NOT want a second node on the network claiming the same name/numbers.

Anyway, the tricky part are probably the device and be device names
- Same network interface? EWxx vs EIxx
- Same hard-drive? DRxx vs DKxx ?

A bit more or less memory, or an CPU more or less will not make a functional difference but may require node specific tuning.

A typically a VMS configuration is NOT dependent on Serial numbers or MAC addresses, but your application might be.

In fact, in light of the earlier warning, you MIGHT want to construct a dependency in the boot process on a known configuration feature or environment variable (firmware!). Thus the system boot procedure could possibly detect whether it is 'home' or on an alternate config and behave accordingly.
Sorry, I haven't thought this through in enough detail to provide you with details.

Hein.

[Peter_364]

Peter,

That's partly depending on how your storage is attached, if it's "local" SCSI-storage you could do as others have mentioned, but if you got fibre-channel storage you could set up boot-path to the very same system disk as the original ES40 uses, but never boot it as the original is up and running.
And you could also configure the backup-machine with the same exact hardware and number of ethernet interfaces and connect them the same so the backup machine will match these like the original if and when it has to replace the other.

[Wim Van den Wyngaert]

None of my applications could do something with such a backup system. I hope you have something to do the roll forward of the data (replay all changes since backup). Or have applications without data.

fwiw

Wim

[Peter Clarke]

Ok i have now managed to do this seem to have a few errors though.
How do i change the hostname and ip address of the server?

Thanks

[Robert Gezelter]

Peter,

I do not see which TCP/IP stack is in use. If it is a recent version of TCP, then the TCPIP$CONFIG.COM command procedure. One can also do this change manually, but it is easier and simpler to use the command procedure.

- Bob Gezelter, http://www.rlgsc.com

[Robert Gezelter]

Peter,

It will also likely be necessary for the DECnet address/name and the SCSSYSTEMID/SCSNODE.

- Bob Gezelter, http://www.rlgsc.com

[Volker Halle]

Peter,

the OpenVMS FAQ has a chapter about this:

5.7 How do I change the node name of an OpenVMS System?

http://h71000.www7.hp.com/faq/vmsfaq.txt

Volker.

[Hoff]

The copy of the OpenVMS FAQ at HP is two editions back or so.

The "most recent" FAQ update is here:

http://www.hoffmanlabs.com/vmsfaq

For this question, also look here:

http://64.223.189.234/node/589

for "Changing the Node Name of an OpenVMS Host".

[Robert_Boyd]

I didn't see anybody mention this suggestion yet: build a 2nd root on the system disk and pre-configure it as if it's going to be a cluster member. Then when you boot the copy, boot from the 2nd root into conversational boot mode and disable clustering. Then you can bring the system on up and finish configuring it. This way you still have the original configuration available for a quick reboot to the 1st root.

Robert

[Phil.Howell]

If you have a mirrored system disk and you are able to shutdown, then just replace the mirror with a new disk, and use it as the system disk on the new machine - isolate it before booting and change the network configuration from the console.
Phil

[DECxchange]

If you intend to make this a drop in replacement, copying the system disk and the other discs will work. But you will probably need to sue the VMS distribution CD to "clone" the system disc. If you boot the CD, it brings up a menu and that is one of the choices, to make an image backup of a disc.

If you want this computer to be available on the network, you will have to change DECnet addresses (sys$manager:net$configure.com), IP addresses (tcpip$config), modparams.dat, maybe licenses, etc.

If you want to make this an application failover / backup, you might get away with changing some logical names and so forth. So it depends on how far you want to take this. If I had a complete spare like that, I would probably want to VMScluster it and setup the application to run on either machine, if one or the other broke down. That is a lot of work though and perhaps a drop in replacement would be the easiest route, at least initially.

[Peter Clarke]

Ok i changed the hostname and the ip address on the interface but when i rebooted the system it came up with the old address and hostname.
Guess i need to change some startup / parameter files anybody know which ones??

Thanks

[Martin Vorlaender]

The procedure to change a nodename (and SCS ID) is laid out in the VMS FAQ at
http://64.223.189.234/vmsfaq/vmsfaq_007.html#mgmt9

cu,
Martin

[Hoff]

What to change?

For this question, look here:

http://64.223.189.234/node/589

[Jan van den Ende]

Peter,

from your Forum Profile:


I have assigned points to 193 of 248 responses to my questions.

Most of those date back to 2004!

Maybe you can find some time to do some assigning?

http://forums1.itrc.hp.com/service/forums/helptips.do?#33

Mind, I do NOT say you necessarily need to give lots of points. It is fully up to _YOU_ to decide how many. If you consider an answer is not deserving any points, you can also assign 0 ( = zero ) points, and then that answer will no longer be counted as unassigned.
Consider, that every poster took at least the trouble of posting for you!

To easily find your streams with unassigned points, click your own name somewhere.
This will bring up your profile.
Near the bottom of that page, under the caption "My Question(s)" you will find "questions or topics with unassigned points " Clicking that will give all, and only, your questions that still have unassigned postings.
If you have closed some of those streams, you must "Reopen" them to "Submit points". (After which you can "Close" again)

Do not forget to explicitly activate "Submit points", or your effort gets lost again!!

Thanks on behalf of your Forum colleagues.

PS. - nothing personal in this. I try to post it to everyone with this kind of assignment ratio in this forum. If you have received a posting like this before - please do not take offence - none is intended!

PPS. - Zero points for this.

Proost.

Have one on me.

jpe

[Peter Clarke]

Right i have finally got this server up and running thanks for all your help everyone but have now gone to login and get this message below anyone know how i sort this licensing issue:

*E* Licensing: PH8-ALPHA-RUNTIME PH8-ALPHA-RUNTIME-W-REPORTING
PH8-ALPHA-DEVELOPMENT PH8-TEMPORARY PH8-DEMO Cannot find license file (-1,73:6)
no such device or address

Thanks
Peter

[John Gillings]

Peter,

>*E* Licensing: PH8-ALPHA-RUNTIME PH8-ALPHA-RUNTIME-W-REPORTING
>PH8-ALPHA-DEVELOPMENT PH8-TEMPORARY PH8-DEMO Cannot find license file (-1,73:6)
>no such device or address

That's not a normal OpenVMS licensing error.

My guess is it's from 3rd party software, possibly called "PH8"? and the license mechanism is expecting to see a device or file which doesn't exist on your new system.

You'll need to contact the vendor to determine how their licensing works (and, indeed, if you're legally allowed to run their software on a backup system).

[Phil.Howell · Accepted solution]

cognos powerhouse?
have a look for license.dat under
[cognos.license]

[Peter Clarke]

Yes i have the license.dat file on that directory.
Do you know what i need to change to get it to work on the new machine all i can see is a server name that i have changed but still doesn't work...

[Peter Clarke]

Its working!! thanks Phil i had missed a logical that that pointed to the license directory...