Disaster Recovery

[sloughman]

Hello

 

I have a meeting tomorrow with the various department heads etc, about disaster recovery. This is the second school I have worked in and I have never had an official disaster recovery type meeting before. This will cover all aspects concerning the school. From my point of view what sort of things should I /we think about (IT wise). Other than the daily back ups of all servers , which is done what else do I need to do/think about /be concerned with ?

 

Has anyone actually had a disaster that they can give useful tips from ?

 

Any info gratefully received.

[Debbie99]

Not in a school, but have had a disaster - we were lucky that the servers stayed up but we could not access the site for weeks.

 

Good that you are taking backups BUT

How often do you test your restores?

Have you ever restored a server?

Where is your backup server in relation to your other servers?

Do you keep backups off-site or in a fireproof safe?

What sort of downtime is acceptable - Risk vs. Probability

 

Think about scenarios.

 

One of your server rooms has blown up:

Can you access your backups?

Can you do a restore - e.g. is your tape drive accessible?

What can you restore to?

What loss in performance is OK - e.g. Everyone using the one domain controller?

What is the most important server/data - probably SIMS, then what?

What timescales are acceptable?

If minimal downtime is needed - how much is it worth, this is your budget.

 

Work out what you would do in each scenario.

A hard drive fails in a server -

A server fails -

A server room fails -

The entire school burns down -

[pete]

Ask "how much money do you have allocated to this project?" at the start of the meeting. Then if they tear off down the "all the things available, all the time", break it to them gently that'll cost more than the school pull in per year in funding (there's no such thing as 100% guaranteed uptime).

 

Read this:

 

The test of a true DR plan: my building flooded. - Ars Technica OpenForum

 

The above isn't me, though we did get flooded in June 2012 (about half the school). Oh, and we got hit by lightning a few years ago (took out a DC, 8 switches totally, misc ports on other switches, random desktops & CCTV).

 

• Know what hardware and software you have (serials etc), where it is, who supplies it (and parts for it) & what it does.
  
• Talk to suppliers about turnaround times in the event of disaster.
  
• Document service dependencies (Server A provides authentication for services 1,2,3,4 and relies on Server B which relies on......) So that if someone says "Bob, we need to get access to "X", you know how to make that happen.
  
• Have your documentation available in multiple locations. And your install media.
  
• If you think something might be damaged, replace it because otherwise it'll die later when the insurance claim has been finalised.
  
• Appoint a member of SLT/senior management whose sole purpose in a disaster is to handle queries/rugby-tackle "X isn't working, when will it be working" people so that the people fixing stuff can do so.
  
• Ensure people get fed, watered and sufficient sleep.
  

Edited January 22, 2014 by pete

[rad]

I restored from our Christmas backup tapes today, I'm happy when it works as it reassures me IF.....

[FN-GM]

A clue is in the name. Disaster Recovery. How long will it take you to recover from a disaster? How long to restore backups? Re-setup switches etc.

[SpuffMonkey]

In this thread http://www.edugeek.net/forums/how-do-you-do/115130-writing-disaster-recovery-plan.html I posted a copy of a DR plan I wrote some time ago - you might find it useful to have a look as it may prompt you to think about how to push things along

[Mr_Jiminy]

Hello

 

I have a meeting tomorrow with the various department heads etc, about disaster recovery. This is the second school I have worked in and I have never had an official disaster recovery type meeting before. This will cover all aspects concerning the school. From my point of view what sort of things should I /we think about (IT wise). Other than the daily back ups of all servers , which is done what else do I need to do/think about /be concerned with ?

 

Has anyone actually had a disaster that they can give useful tips from ?

 

Any info gratefully received.

 

What have you currently got in place in relation to disaster recovery and business continuity?

 

Ask them how quick they want there data back and work from there - of course it entirely depends on what you have in place, then how you improve upon that.