War damage to data centers exceeded what AWS services are designed to withstand.
Half a year after Iranian drone strikes knocked out multiple Amazon data centers, the US tech company has acknowledged the permanent loss of some customer data that was hosted in Bahrain and the United Arab Emirates.
Amazon Web Services was “unable to restore access to the resources and data” hosted in some of the war-damaged data centers, according to an AWS dashboard update posted on September 15. The development was first reported by Reuters and suggests that the Iranian strikes inflicted catastrophic damage on the data centers.
Customer data was irretrievably lost in one of three AWS availability zones in the United Arab Emirates region, specifically the mec1-az2 availability zone. Each availability zone is serviced by one or more Amazon data centers.
Anyone in IT who has ever been near a backup or had to do a DR exercise knows full well not to have a backup in the same geographic area as your primary workload. Usually it’s for natural disasters, but it very much applies to war too.
This is just as much, if not more, on the companies that didn’t have geographicly dispersed backups.
The 3 2 1 rule.
3 copies of the data
At least 2 different kinds of media
At least 1 needs to be off site.
A not insignificant part of my first azure cert covered different redundancy options and pricing. If the data was that critical, the tenants would/should have mirrored it to another region.
Depending on the data requirements they may not have been legally allowed to move it to another region. Medical information or banking data usually has an on shore requirement. Sometimes funds have issues with storing Information overseas esspesully if there is concerns about other governments gaining access.
“yes but it’s not in the budget this year”
I think pretty much every software dev has used AWS at some time, and when you set up new services, it specifically warns you to use multiple geographic areas for your services. It doesn’t say “war” or something, only that it’s good for redundancy, IIRC.
So anyway, yes, I follow instructions, but I’m always surprised that you have software people and systems admins who can’t even follow basic instructions. I don’t know what percentage of people it is, but it feels like over 50% who can’t follow even the simplest instructions.
I’m not surprised at all. It costs more to use more regions, both due to literal cost and in time to set it up.
The extra time to set up in more regions is pretty trivial, at least in my experience. It’s either an option when setting up the service, or it’s something that you need to cover in any case while documenting how to set up the service. But I suspect the devs I’m complaining don’t read basic instructions also don’t write basic instructions.
As for the monetary cost, that’s a business decision. The devs I am complaining about are not making business decisions. They’re just ignoring basic instructions.
I haven’t used aws for ages but “an option when setting up the service” does not encompass my recollection of it: you can set up (say) S3 in multiple regions, but that doesn’t mean your application can actually make use of multiple regions. It doesn’t mean your vlans actually enable anything to talk between the regions.
I haven’t used it for a while, either, but it was just an option sitting there when using the AWS UI to set up RDS, IIRC.
Right, but how is your application going to handle it if you enable it?
You have to set up your application to use the service, regardless, so I would say that you just have to follow the instructions.
I am not talking about fixing after a service has been set up for only one region, but about following the instructions when first setting it up.
Yes, but the application setup is more complicated if it can talk to more than one region.