tru 64 backup and revcovery

[victory_1]

Hello

I am new to tru 64 OS ver 5.1 .I need detailed documentation and web sites for backup and recovery in a DR scenario.

Thanks

VIJAY

[Mulgund]

There are some good topics on archiving
in these docs.

System Admin. Guide:
http://h30097.www3.hp.com/docs/base_doc/DOCUMENTATION/V51B_HTML/ARH9FETE/TITLE.HTM

AdVFS Admin guide.
http://h30097.www3.hp.com/docs/base_doc/DOCUMENTATION/V51B_HTML/ARH96DTE/TITLE.HTM

LSM guide:
http://h30097.www3.hp.com/docs/base_doc/DOCUMENTATION/V51B_HTML/ARH9BDTE/TITLE.HTM

I haven't seen any doc. specific to disaster recovery in Tru64.

Hope this helps.
-Srinivas.

[Han Pilmeyer]

If the first reply doesn't help you, then perhaps you should describe in more detail what you mean by "backup and recovery in a DR scenario". There are quite a few different things that could apply to such a description.

[victory_1]

Well going through the first reply , I have got

info on how backup and restore is done using

vdump and vrestore. I want more details

regarding how when a system crashes , file

system becomes corrupt ,devices fail to get

identified , Is there any stepwise dr backup

and recovery process that is followed like

making a boot tape,etc.

Any good documentation on the same will be highly appreciated

Thanks

VIJAY

[Mulgund · Accepted solution]

Sec. 9.7.5 Restoring or Duplicating a System (Root) Disk of Sys. admin guide details about root disk recovery.

AdVFS Admin guide describes this:
6 Troubleshooting
* Recovery of a single user file (Section 6.1)
* System recovery (Section 6.2)
* Recovery and restoration procedures for a number of common problems (Section 6.3)
* Preventative maintenance strategies (Section 6.4)

Sec 9.10 : Creating a Standalone System Kernel on Tape using bttape utility.


Hope this helps.
-Srinivas

[victory_1]

Hi Mulgund

Thanks a lot . I think that is what I was

looking for and should be enough info .

Thanks

VIJAY

[victory_1]

Hi Mulgund

I went through this and the document becomes

quite huge. Can you please help me with a brief

document which just briefly tells the steps

with the commands to do a recovery

You can send the doc at my email id :vrags@hotmail.com

Thanks

VIJAY RAGHAVAN

[Johan Brusche]

Vijay,

You migth find some useable ideas in the best practice section:
http://h30097.www3.hp.com/docs/best_practices/sys_bps.html

The basic steps in a DR situation (ie data replication) are:

freezefs (to stabilize fs metadata)
Stop replication
thawfs
Make replica available to OS.
mount -o dual filesystem on replica
Make backup with 'intelligent' backup software like OpenView Dataprotect or Legato Networker. These soft packages know how to handle a jukebox and archive indexes.
vdump has no such functionality.

__ Johan.

[victory_1]

Hello all

I have made a small doc which covers all apects

as given by your suggestions.

Please review this doc and give your valuable

feedback

Thanks

VIJAY

[victory_1]

Thanks for all help and all replies.But the client whom I am working with does not want to go for a tape restore solution.

Please help me with a brief doc which explains
about system cloning capabilities from which the OS can be recovered.

I would appreciate an early reply as we need to

give this to our client at the earliest

Thanks

VIJAY

[Han Pilmeyer]

That's why I asked you to describe your question a bit better. There are too many options.

Looks like you're looking for the Ron Urban system / cluster cloning scripts. They copy the system disk(s) to another LUN(s). After that all you need to do to active the clone is to change the boot path.

Look here: http://www.unix-wissen.de/Tru64/clone-cluster.html

[victory_1]

Han

Thanks for the reply. I would appreciate if you

can go through this doc and let me know if this

enough for a client who wants to use system

cloning capabilities in 5.1B of TRU64.

I also would like to know how you can restore

back after cloning.

Thanks

[Han Pilmeyer]

Vijay,

I'm afraid we will continue to run around in circles until you clearly explain what the problem is that you are trying to solve.

The document you attached describes two completely different things. Both could apply to what you're trying to accomplish. I can think of several other things that could be just as applicable. On the other hand, non of this may apply to the situation. We just can't tell until we know what the problem really is.

[victory_1]

Hi Han

Thanks for the reply. The requiremnet is

about system cloning capabilities in 5.1B of

TRU64.

The client would basically like to go for a

cloning solution as they do not want to go

for a tape restore solution.

They want to keep a cloned image of all the

root filesystems and recover the same when a

system crash happens.


The doc which i sent expalins cloning process

in tru 64 but I am still wondering as to what

is the solution if we want to recover using

these clone filesets.

I am sorry if this is going in circles but

this is my exact requirement.

I would appreciate if you can give me some

feedback on this .

Thanks

VIJAY RAGHAVAN

[Han Pilmeyer]

Cloned file sets won't help you in that case, They reside within the same domain as the original. Therefor if anything happens to the domain, you will lose both.

The cloning you mentioned in the document could help you, but is probably even more cumbersome than doing a tape restore. The purpose of this cloning is to install multiple (new) systems that are almost identical. Unless you script every change that you make on the system disks and then re-apply those after you install the system again (using the cloned installation process), you will lose those customizations.

Depending on the storage that you use, you may be able to use hardware clones (business copies) of your system disks. However these have different WWID's and since Tru64 UNIX identifies disks by their WWID, you will have to come up with a process to account for that (which is very difficult in clusters).

The cloning scripts from Rob Urban that I pointed you to earlier, might do exactly what you want. Each time you run those scripts they create a copy of the system disk on another set of disks. The script allow you to make some simple changes to the system disk (such as partition sizes, quorum and votes in a cluster, etc.). Typically people run those scripts before they do major work on the system disks (such as installing a patch kit) or they run them nightly from crontab. When you need to recover, all you have to do is to boot from the cloned system disks. It does require that you have a spare set of system disks, which on modern storage arrays is usually not a big deal.

It's also a good idea to run sys_check after every significant change on the system disk as sys_check saves all important configuration information (including LSM meta data). You should then keep the sys_check output on another system, so that you can access it when you need to do a recovery.

[victory_1]

I have got all the info required.

Thanks to all