FTP server backups

@Patrick:  you are making assumptions about the situation that were not stated.  (ie: Core with Server 2008 & BMR)

OK, so maybe your automated deployment takes 20 minutes.  But one must assume that you have equivalent server hardware available to replace the one that just crashed.  I have BMR available and unless there is matching hardware in storage, BMR will be (and has been) nothing but issues.

Risk Assessment:
1.  Value of the data stored on the server?
--- (ZERO, we know this from the original post)
2.  Cost to the business per hour, if service is offline?
--- ?
3.  Cost to the clients per hour, if service is offline?
--- ?
4.  How frequent is the service hit?  once a min, once an hour, once a day, etc?
--- ?
5.  As Patrick mentioned, what is the acceptable service downtime?
--- ?

If this is a business critical service, than having a single server is the issue and that is dangerous to the business.  The solution there is not a BMR task, it would be to build the service to be highly available.  That would reduce the BCP risk.

Business Continuity Planning is about keeping the company functioning and planning to prevent service disruption.  That is different from Disaster Recovery.  This question (as I read it) is about DR and the procedures that can be used to recover the event of an outage.

Stated in the question is the fact that the admins there stated that a fresh install would be faster.  I'll assume that the admin guys there know their environment better than we currently do.

My opinion stands as presented.  If there is no long term data stored on the server and it only provides the transfer service via FTP, I see little value in backing the service up.  It can be rebuilt cleaner/faster than thru a restore.

Dan

@Dan, i respect your opinion but until it is not clear what is acceptable outage......a production machine, for me, is always important.
About the core stuff i must explain myself, core Windows in this case means a simple install of Windows server 2008r2 (GUI) fully patched but nothing else besides a little ftp config.
This tells me, plus minus, 40gb in totall maximum (20-30 minutes to backup)

I would make Once a week a full metal backup (in my case with Acronis).
In case the server dies in the middle of the week i can use any iron available that has almost equilavent hardware and i can restore my latest backup in 20 minutes without no hassle at all. Since there is no important data i can afford to do backups Once a week or less.

Then my assumption is. 20 minutes is way better then hours. Thats 20 minutes for even a new engineer that does not know the ftp setup well.
Offcourse i have no doubt his engineers know their enviroment better, i am almost just as convinced that when admin take disaster recovery so lightly they are more or less arrogant or ignorant.

Guess we feedback from the OP.

Dan