- Sunday, 22 November 2009
- INQ Mobile
- RSS
If I had all the money I've spent on drink, I'd spend it on drink. - Tame Alien
Ballmer admits the Sidekick fiasco is not good
DELIGHTFULLY UNDERSTATED Microsoft CEO Steve Ballmer has admitted that the company's recent Sidekick data loss episode is "not good."
Cornered by Network World, Ballmer said he was sure data lost during an upgrade of Microsoft's 'Danger' system eventually would be recovered.
Ballmer said now Microsoft was going to have to work out a way of explaining why this sort of thing would never happen on the Vole's much hyped cloud services.
Ballmer said, "It is something we are going to have to address and explain to customers our method and process and quality approach and what went wrong in that case and how we are making sure that it does not happen again."
Microsoft will have to earn back the trust of Sidekick users, some of whom reportedly have filed lawsuits, he said.
He actually played down the amount of data that was lost in the cock-up. Ballmer said that it was not clear that there was data loss, although the Vole thought there had been. However, although he couldn't be sure whether any data was actually lost, he was certain that Microsoft would get all user data back.
But while Steve seems to be admitting the problem, according to Roughly Drafted, Microsoft is pointing fingers at its competitors. The system involved in the crash was being run on Sun hardware and an Oracle database was being used, which the Vole was running to fulfill contractual obligations to T-Mobile for the Sidekick project.
The Vole seems to be telling the world plus dog that it was running a project it didn't want and its rivals' gear and software let it down.
That would all be well and good but for the fact that the Sun-Oracle combo is not exactly rocket science and is popular among those who manage very large, mission critical corporate databases.
Instead of following proper system upgrade procedures, apparently a senior Microsoft manager ordered technicians performing scheduled maintenance to work without performing backups first, in order to save time and money.
Apparently they were two days into a six-day backup procedure when orders came from the Vole manager telling them to stop the backup and start the SAN upgrade.
The Vole claimed it had assurances from Hitachi that a backup wasn't necessary but all of the engineers thought this was daft. They reportedly had a backup from a couple of months ago, but only had enough SAN space for a single backup. Because they started a new backup, they had to remove the old one first.
After the SAN upgrade, disks started 'disappearing' so the Oracle software freaked out and started trying to recover, which just made the damage worse.
As Steve Ballmer said, "Not good". µ
Share this:
ShareBusiness Intelligence Software
Customer Relationship Management (CRM)
Enterprise Accounting Software
Enterprise Resource Planning (ERP)
Enterprise Systems Management
Even in being critical of the Almighty Vole Nick shows his bias one can only imagine if Apple had lost this data what Nick's tirade would have been like.
Microsoft continues to behave like a Vole. Don't admit anything. Try hard to blame it on someone else.
I'm counting all the excuses Microsoft has used about the data loss debacle:
1. Blamed it on the Sun hardware and Oracle database, claiming the debacle happened because "it was not Microsoft technology".
2. Blamed it on a company called Danger, which Vole ate 18 months before. How long does it take before Microsoft takes responsibility?
3. Blame Hitachi. What a great excuse: "The Hitachi guy told us that back-ups are for wimps, and not needed. We believed him."
4. Blame T-Mobile customers' wacky imaagination. Now Ballmer disputes there was ever a data loss at all. That's akin to saying the T-Mobile customers just imagined there was a data loss.
If Microsoft doesn't think it was its own fault, how will it ever ensure it doesn't happen again? It doesn't instil much confidence in Microsoft's Azure cloud service if it doesn't admit to cock-ups. And who in their right mind would use the Windows Mobile 'MyPhone' back-up service? (Who in their right mind would use Windows Mobile at all?)
The Vole needs to come clean before it can earn our trust.
Jip, good thing it was not the evil apple company!
Jip, good thing it was not the evil apple company!
I love the statment about the auto recovering Oracle database. Im an Oracle DBA(among other things) and this simply is not true. The Vole can't even admit the admin tried to recover the databsae. How silly just admit you F$%ked up.
"Ballmer single-handledly rescues Sidekick users' data!" (paraphrasing the New York Times).
LOL if I had the time to bother, I'd search for .Mac coverage on Inq and compare/contrast Nick's coverage with this.. .Mac was bad but this was far, inexcusably worse, but I bet the vitriol and snark about .Mac was typically insufferably dopey.
Nick is a tool, useful only for driving Mac fanboy pageviews. If inq had an author-ignore feature he'd be my only entry :p
(Frankly, with Mike, Charlie and Fudo gone, why do I even keep coming back? Habit I suppose :p)
There are several articles on this item that blame 'Cloud' computing. Based on what we now know about the back-end infrastructure, Sun HW and Oracle SW, blaming Cloud computing is a little disingenuous. The infrastructure was traditional and the failure was violating standard backup best practices. Microsoft should have moved this over to Azure but the SW was probably not ready for distributed architectures.
The real issue is not Cloud. The issue is not following best practices or internal processes.
Dear Steve,
You just need to look at 2 names on the danger team:Mark Fisher and Jamie Rosenberg
Keep them away from windows mobile. Check out their bios on linkedin.
Yet another decision made by those who know the cost of everything, and the value of nothing.
To be honest, backups are not given the priority they should be, because people have been lulled into a false sense of security, thinking that everything "just works", especially upper or project management staff.
Many companies were burned in the 80s and 90s because of failure to back up their data, and it would seem that memories are very short - because these lessons are, it seems, still being learned by today's management.
"They reportedly had a backup from a couple of months ago" - Can can you believe this? Their standard practice was to only backup every couple of months.
Never, Ever, trust the Vole with your data. Who cares if this isn't technically their cloud system. If a system containing customer data is allowed to run for MONTHS without backup then something must be very rotten in the management of Microsoft.
And how about the whole having to delete the last backup before a new one can be started? Sure it's all on a redundant SAN, but I think they've just proven yet again that redundancy is not backup. DON'T TRUST THEM, EVER. It's just too big a company and the management too rotten - eventually some of the internal groups will start cutting corners again and this will happen again.
I'm quite sure it will never happen at all - to me anyway. I'll never trust a cloud with my data.
Not unless I have my own backup, that is.
It amazes me how many recovery scenarios occur without a recent, tested, backup. Hitachi may have said it would work but they didn't pull the trigger. The engineers must have been pissed because then they get to clean up while Balmer talks smack.
James, unfortunately, the avg person doesn't know that and he is trying to convince everyday people it wasn't M$'s fault; not people that actually have a clue.