Does it make sense?

[key]

Hello,

I had 3 physical servers (Windows 2008R2) that connected to SAN storage for data access. The OS "C" drive is on physical server while the data "D" and "E" drive are pointing to the storage. I do have backup running everyday to backup a complete system for C, D and E drives.

To complied with our IT policy and auditors, twice a year, I did performed a system restored from the backup to our test environment to make sure the backup is good. I also had users sign-off to confirmed that all the data are corrected and up to date (to the time of backup)

Today, I loss all the data in the storage due to SAN configuration got "factory reset". (note: all data loss is about 600GB)

Our IT goal is to make all 3 server up and running as soon as possible.

Here come with the decision between 2 options:

1. Restore the servers from backup according our IT procedure (that we performed twice a year) which can bring all the servers up and running for less than 12 hours.
2. Got a quote from consultant that cost $300K ($300000) for:  

      * Buying new storage with new hard drives

      * Buying new "hardware" and "software"

      * Consultant fee

      * 5 days down time

Since we only lost the data and there is no issue with our hardware, does it make sense for consultant purchase all kind of hardware, software and 5 days downtime? or should I just perform a system restore from our backup?

Thank you in advance

Key

 

[Monoman]

I would do #1 but if my job depended on the recovery, and you're not 100% confident I would go for #2.

 

[BLinux]

if your goal is quickest recovery, then #1 and keep #2 as backup. You will know in 12 hours or less whether you need to do #2.

alternatively, I would consider getting a 3rd opinion from another consultant that can help you with your recovery goal without such an expensive option as #2. 600GB is not a lot of data, and why the need for all new hardware? that mostly doesn't make sense unless there's other context the consultant is aware of that isn't revealed in your post. there should be an option #3.

btw, for example, if your consultant knows that your SAN hardware is really fragile and that's why the configuration got lost and data got lost, then it might make more sense why he/she is proposing to buy a lot of new hardware. but you didn't tell about the root cause to why the SAN got "factory reset"? Did someone make a mistake? (human error) or is the hardware not healthy and something else happened?

 

[key]

if your goal is quickest recovery, then #1 and keep #2 as backup. You will know in 12 hours or less whether you need to do #2.

alternatively, I would consider getting a 3rd opinion from another consultant that can help you with your recovery goal without such an expensive option as #2. 600GB is not a lot of data, and why the need for all new hardware? that mostly doesn't make sense unless there's other context the consultant is aware of that isn't revealed in your post. there should be an option #3.

btw, for example, if your consultant knows that your SAN hardware is really fragile and that's why the configuration got lost and data got lost, then it might make more sense why he/she is proposing to buy a lot of new hardware. but you didn't tell about the root cause to why the SAN got "factory reset"? Did someone make a mistake? (human error) or is the hardware not healthy and something else happened?

Thanks Blinux for your response.

Option #1 is what I'm thinking as I did do the system restore twice a year to my test environment and users did verified the restore does works.

The reason the SAN got "factory reset" due unauthorized access from former IT.
This infrastructure (included Servers and SAN) was build by outside consultant.

 

[BLinux]

Thanks Blinux for your response.

Option #1 is what I'm thinking as I did do the system restore twice a year to my test environment and users did verified the restore does works.

The reason the SAN got "factory reset" due unauthorized access from former IT.
This infrastructure (included Servers and SAN) was build by outside consultant.

Ok. so if the problem was due to human error, there is no reason to buy new hardware.

i would fire that consultant that proposed #2 and find someone else.

 

[Connorise]

1. Restore the servers from backup according our IT procedure (that we performed twice a year) which can bring all the servers up and running for less than 12 hours.

This.

And I do agree with this "I would fire that consultant that proposed #2 and find someone else"

 

[Connorise]

What about a solution:
I believe HCI would be great. You will get high availability and additional redundancy.
In any case to do not lose your data you better think about redundancy (2 server in HA will provide you with it).
You can try something like vSAN from VMware or StarWind (it is free and pretty easy to use)

 

[DavidRa]

Elephant in the room - how is that consultant going to get the hardware purchased, delivered, installed, built, tested and commissioned in zero time, which would be the timeframe to have the existing hardware ready to go? If you take only option 2, you're committing to a week* of downtime. That seems completely bass-ackwards to me, especially since you have a working DR process already.

* You're dreaming, a month.

 

[EffrafaxOfWug]

Not that I'm cynical or anything but the consultant must either be idiotic or an outright conman.

Elephant in the room - how is that consultant going to get the hardware purchased, delivered, installed, built, tested and commissioned in zero time

Maybe he has leftover hardware from another client who wised up and told him where to shovel it and he's desperate to offload it? But yeah, as a DR procedure, option 2 above makes about as much sense as sodium buttplugs.

 

[Myth]

We sell SAN servers here. I've never lost data on a server due to a "factory reset" our data is stored on an HW raid controller via a windows 2016 physical server. We can do an automatic redundant offline backup. We can do this for way less than the quote above.