Disaster Recover Plan

[rama1712]

We have been asked to write our Disaster Recover Plan, Does anyone have a plan that we can have a look at an maybe with permisson customize it for our use.

 

If you have PM and I will send you my Email Address to send it to

 

Stuart

[Ric_]

eek... I keep meaning to write one but it's one of those things you put off because it is a PITA.

 

Don't forget that you need to specify the 'disaster', determine what must be done with instructions on how to do it and include any software or utils required. You will potentially need several scenarios ranging from disk failure to complete theft/destruction of your server room.

 

It might also be an idea to include a section on how damage potential has been limited (e.g. not keeping all your servers in one room).

 

It will take you a while to write so good luck!

[pete]

Start here: http://www.sans.org/reading_room/ they have a disaster recovery section.

 

If you backup to tape, ensure you can actually still get (or know of) another tape drive that can read them. Always fun to discover you can't.

 

and include lightning strikes into your plan. It will cost you 2+ ports per affected switch and you'll probably lose a couple completely. Server NICs may need replacing (or mobos if they're integrated). No, having them on a UPS doesn't save them if they get bitten by the EMP or the strike comes through your Cat5.

[Netman]

Here's a copy of ours (with certain details removed!)

It's not specific on types of disaster but just general procedures.

Anyone may use and change it as they require...

ict_disaster_recovery_plan_example.doc

[plexer]

Not so much a plan as a crib sheet but thanks for sharing it with us. Not that we have one at all at the moment. A DR plan needs to be detailed covering steps and resources etc... needed to recover the system.

 

Probably written so as to be followable by a numpty. Death of network manager is probably a disaster have you planned for that?

 

Veritas used to have lots of information on DR I attended a DR seminar at the HQ once. Obviously very backupexec centric but was good all the same.

 

Ben

[plexer]

Found some examples of DR plans:

 

http://www.drj.com/new2dr/samples.htm

 

Ben

[SpuffMonkey]

Probably written so as to be followable by a numpty. Death of network manager is probably a disaster have you planned for that?

 

I don't know that you could ever write a plan that detailed. I think its better to have defined milestones and targets - so that expertise can be bought in if required.

 

Alas - I think that if I was dead under a pile of smoking servers, worrying about the DR plan will be low on my list...

[rama1712]

It Seems I have open a can of Worms with a DR plan,

 

We have been asked to produce a plan ready for Auditors In January I was told 1 page would do and to generalliase I might stretch to 2 pages

 

I will let you all know how i get on and post it when done.

 

Stuart

[plexer]

A summary is not a plan though.

 

Ben

[Netman]

Death of network manager is probably a disaster have you planned for that?

 

yeah - it's called the 'buy a new network manager plan'

[GrumbleDook]

When we had the Becta audit last academic year we held our hands up and say we had no plan but a series of ideas about how we could get things going quickly based on some existing good practice.

 

We were actually marked up for admitting we did that rather than push out a 3 pages doc that we called a Disaster Recovery Plan.

 

The DRP we had for the garrison I was based at reached 7 volumes ... and that was not including referring to existing publications.

[john]

Disaster recovery plan, whats one of them has been the case in the places I have worked . I think I will write one though for the places I do bits of work for as it would be most useful for them, I already write a mini IT handbook for them with things like warranty contacts, serial numbers, locations, Ink / Toner Cartridge Part Numbers etc... little things I know but very useful for little schools who don't have a full time IT dude to phoen.

[Ric_]

I know that Lancaster Uni undertook a massive DR project and ended up creating little boxes for every piece of hardware. Inside each box was the software media, serial numbers and documentation for recovering from a 'disaster'.

[john]

@ric - ooh, that sounds sort of similar to what I do, I make a "pack" for each type of workstation and server I have, probably not quite as thorough as they have done, but it has key things like a copy of XP, driver CDs, and a quick paper manual / info page - generally the mainboard diagram labeled to show where things like CMOS reset are etc just in-case its needed to be reset to original for emergency use etc, nothing more special than that though, but in an emergency that will surely be of quick assistance for getting at least a PC working and operational to allow emergency data access to say an MIS etc...

[ITWombat]

eek... I keep meaning to write one but it's one of those things you put off because it is a PITA.

 

Yes, what is the school meant to do if the terminal server farm goes t1t5 up and their local Citrix guru is suffering fom the effects of a most potent curry.