Microsoft: We May Have Recovered Sidekick Users' Lost Data
Maybe those lawsuits over the Sidekick data outage were a little premature. Microsoft, parent company of Sidekick maker Danger Inc, reports that they have recovered most, if not all, of users' lost data. Yay!
Microsoft blamed a system failure for the data loss in the core database and back-up system. The company said it had taken steps to strengthen the stability of the Sidekick service and started a more resilient back-up process.
Over the weekend, T-Mobile and Microsoft initially warned that the recovery of data would be unlikely, but upgraded their prospects on Tuesday.
Microsoft said that an additional update would come on Saturday and provide a timeline and additional details on the recovery.
So the restoration won't be immediate, but the lost data may not be permanently lost. Maybe. That's certainly an improvement over the previous status of "It's never coming back, and here's $100."
UPDATE: Microsoft Says It Has Recovered Lost Sidekick Data [Wall Street Journal]
PREVIOUSLY:
T-Mobile: We Won't Swap Out Your Sidekick For A Different Phone
First Sidekick Data Outage Lawsuits Filed
T-Mobile Sidekick Data Outage Turns Into Epic Customer Data Fail
Post a comment
Comments:
Of course, MS will pass the buck, and say the data failure was due to an issue with their hardware, and pass the blame to the hardware manufacturer.
Let's see if the Sidekick brand has been destroyed like some were predicting.
After this, T-Mobile should be distributing free software that creates local backups of Sidekick data to it's users.
I wondered why they said it was completely irrecoverable. Unless you actually wipe over the data (takes a very long time) its still sitting there on the hard drives, no matter how much you screw it up. Might take a ton of money, but it still be recoverable.
So, first they said the data would come back, then they said it wouldn't and began crediting people, now they say it will - what happens to those credits?
@W10002: I have a feeling if it was "Apple, parent company of Sidekick maker Danger Inc.," then it'd be lots of applause for doing the right thing. See: NVidia's Mac chipset problems.
Since it's Microsoft, obviously they're just passing the buck because it couldn't possibly be someone else's fault other than theirs.
@nbs2: Publicity motivated some due diligence theatre. It's likely they aren't planning on paying for both recovery and service outage credits.
The uncertainty and mess plays into MS' hands regarding discrediting the bastard child of the family, at least to those who've already sold their souls. "It was a Danger architecture problem, microsoft saved the day, but you really should move along to our windows mobile (gag) platform for business-critical apps." I just threw up a little.
Give then length of time it took to get the "we can probably recover most of the data" answer, I'm curious as to whether or not it took a while for high-level Microsofties to realize how big the lawsuit over their utter failure to live up to the SLA in T-Mobile's contract was going to be and then light a fire under some systems admins and data-recovery specialists.
"Sir or Madam, we believe we may have recovered your missing data. Can you tell us what it looks like?" Heheheh...
"they have recovered most, if not all, of users' lost data"
"So the restoration won't be immediate"
I think this is BS. Microsoft right now is screwed. This loss of data destroys Microsoft's desire to move us to cloud computing. They had to come up with something and do it fast.
Did they give everyone back their data? Nope.
Did they find all the data? Nope.
Did they find some of the data? Maybe.
This is nothing but a press release to calm everyone down. The media will report that Microsoft saved the day and that's the impression most people will have of this catastrophe.
Weeks will go by and a few users might see their data show up. Most will not. But by then the media will be bored of the story and won't report that this press release was bullshit.
@dohtem: They already do but people don't use it because its to hard to remember to push a couple of buttons.
As an IT professional, working in large scale data centers, it's hard for me to imagine how they could have hosed this so badly. For mission critical data, you have an alternate site and replicate the data from, say, Seattle to Dallas every day. And you don't just replicate it, you save the old copy in Dallas until you are sure the replication has succeeded. And then you back up the Dallas copy to tape. It takes more than just a casual screw up to lose your data.
Oh, unless of course you don't consider the data mission critical . . .
@W10002: I work in IT, and I can say that hardware failure is actually somewhat common. I've seen a situation where there was a hard drive failure, which led to the discovery that the RAID controller had not been functioning properly for the past month (which meant no backups).
Of course, hardware failure isn't an excuse for data loss; it just means that your plan for backups wasn't good enough.
@GearheadGeek: This is exactly what I was thinking. They figured they could get away with the $100 and not bother with the expensive recovery process. Nice try, MS.
I heard that Mr Ballmer did this deliberately.
The logic is thus: if everybody moves their stuff to the cloud, then nobody needs MS Office, or MS Windows. Even though they're working on their own cloud offerings, the real money is in Windows and Office. The cloud is bad for Microsoft.
So MS creates a widely-publicized cloud meltdown, with at least some of the hate redirected to T-Mobile.
After a few days of garment-rending, they suddenly find the data ("Oh, HERE it is, behind the sofa!"), and so their customers are made whole. But everybody is still turned off of cloud computing, and they put away their shiny Chrome browsers and Google Documents and whatnot and go back to Windows and Office.
Hmmmm...
@boshaus:
If the data was on a striped environment where it's written spread across multiple disks, which is done to improve performance, that sort of forensic recovery is next to impossible.
There are plenty of drive recovery places that can handle recovering RAID arrays, which I'm sure this was. I've even imaged 4 drives in a raid 5 and used software to reconstruct it based on 3 good drives. Took 2 days of work... but it recovered it. There's only so many stripe sizes and orders the disks could be in. Just a matter of time.
@KTK1990:
Any enterprise would store it on a raid 5 or better, which allows at least 1 drive to fail and not lose the data. Unless you get 2 or more drives physically mangled, its all there.
@Hawkins:
Aside from the fact that this is a crazy tinfoil hat conspiracy theory... MS is releasing a web-based version of Office.
@SatanicGuinea: Oooooh! Busted!!!
I'd like to add that no corporate backup/disaster recovery strategy is really ever established until it's tested. Just because you've got a ton of stand-by RAID arrays and shelves upon shelves of LTO tapes doesn't mean it's going to make a lick of difference when the shit hits the fan.
What was the figure? 80% of companies that have a major data disaster go out of business within a year?
Seriously, what is so important on a cell phone that people are freaking out about compensation and the like? Boo hoo! I lost my calendar/address book/myspace bookmarks! If this happened to me, yeah I would be pissed and it would be a b*tch to replace, but s--- happens...
@lordargent: It is not a substitue, but many company store their backups on RAID setups.
At my old company, we had a large raid in place which backed up all data localy, then would upload it offsite over night.
Maybe they should patch the firmware on the Sidekicks so they don't delete local data when the remote servers decide to take a shit. But that would have made too much sense to have done from the device's inception.
@SatanicGuinea: raid on its own is not a complete solution... it provides redundancy, but not backups.
backups on their own take time to restore, so they aren't a perfect solution.
a GOOD solution involves redundant hardware (RAID), onsite AND offsite backups, and preferably colocated servers.
@mbz32190:
When you enter the business world and need your real time communication then you'll understand.
They probably found it sitting on a hard drive in some tech's PC.