Installing hpux on san disks

[Emiel van Grinsven_1]

Hi all,

Not sure if this is the right Category but here goes...

I'm interessted to hear your opinions and maybe also experiences with installing the OS on san disks.

I have this in mind so that we can offer a disaster recovery option to clients who don't want to use a cluster or have high availability but still in case of a site disaster want some kind of recovery option.

The basic situation:
we have two data-centers, in both centers we have san storage (in this case EMC symmetrics) that can be linked with SRDF.
What I want is to install the OS on site 1 on the san disks, than in case of site disaster boot a system on site 2 with an alternative bootpath so that it boots from the SRDF disks that were mirrored from site 1.
The hosting system on site 1 is for a prod environment, on site 2 for a test environment.

Questions:
#Can it be done? (hmm, I think I already know this answer)
#What are known problems?
#Specific software? Like veritas volume manager needed? Powerpath? etc?
#Problems with unique system ID (uname -i) and software licences?

Yes, I'm also taking to an EMC enigneer at the same time so. I want to hear the viewpoints of a systems engineer/admin.

Thank you in advance.
Emiel

[harry d brown jr]

We have over two hundred hp servers booting from emc san disk per datacenter (2). We do not srdf our boot disk - but we do srdf our other volumes. the other datacenter has boot disks and special recovery scripts to break the srdf, import the columes and start the applications.

Just make sure you aren't doing any isl hops for your boot disk.

live free or die
harry d brown jr

[Emiel van Grinsven_1]

Why not the boot disks? Performance?
How do you maintain those scripts that you use?

Thanks

[harry d brown jr]

Why not the boot disks? Performance?

*** Nope. Not my decision. I proposed it, but I was over ridden because some people thought it was too complicated. They would have to make both SAN's in each data center identical so that luns matched. Imagine that!! We have two data centers with each capable of running 100%, with each running 50% with failover to the other data center. I call that LAZY not complicated!!!

How do you maintain those scripts that you use?

*** vi

live free or die
harry d brown jr