JournalSpace data loss terminal

Author: Gareth Halfacree
Published: 7th January 2009
Comments (10) Stumble
JournalSpace data loss terminal

Although JournalSpace is gone, the Rhodes' other projects continue - hopefully with an actual backup solution in place.

The folks behind blogging site JournalSpace.com have learned the hard way that RAID is not a substitute for backups, with the news that a disgruntled ex-employee has taken down the site completely.

According to an article on BetaNews, the bloggers-turned-sysadmins behind JournalSpace relied on a dual-disk mirrored RAID system to keep their precious SQL databases safe. Unfortunately – but perhaps not surprisingly – while the RAID mirror would keep the website ticking over in the event of a single disk failure, it is powerless to prevent data loss due to software failure or active sabotage.

It's the latter that the JournalSpace admins believe caused all their data to disappear in the blink of an eye: the admins claim that a “disgruntled member of the Lagomorphics [hosting company] team” was responsible for the data loss, and say that the individual in question “sabotaged some key servers several months ago after he was caught stealing from the company.”

Once the data was gone, the company tried to recover the files from what they, perhaps naively, believed was a 'backup' – only to find that when a file in a mirrored RAID system is gone from one drive, it's naturally gone from the other. Despite sending the disks to a data recovery company, the data was gone for good.

While this is a perfect lesson in why RAID is not a substitute for backups, it's one that the company has learned at a very dear cost: according to the website the company is shutting its doors, with the domain name being offered for sale on eBay. Customers who have lost their journals are pointed at Google's cache as the best way to retrieve at least some of their personal data.

While it's sad to see a company fail – especially as a result of possible vandalism – it's perhaps amazing that JournalSpace was able to last for an entire six year period without a backup procedure in place. Perhaps the founders, Dylan and Holland Rhodes, will get one set up for their remaining on-line ventures – Mac-specialist hosting company Lagomorphics and photo-sharing site Reezle.com.

Did this tale of digital woe have you running to check your backups, or are you safe in the knowledge that your disaster recovery plan is in place? Was this an accident waiting to happen, even without a disgruntled ex-admin in the mix? Share your thoughts over in the forums.
Quote [USRF]Obiwan 7th January 2009, 12:14
Still, I believe if there was not a 'annoying person' that deleted all the stuff, the data would have been relative save if one of the mirror drives failed due to hardware fault. If there was for example a DLT tape backup. The same person could have destroyed all the DLT backup tapes along the deletion of files on the hard drives.
Quote Gareth Halfacree 7th January 2009, 12:28
Quote:
Originally Posted by [USRF]Obiwan
Still, I believe if there was not a 'annoying person' that deleted all the stuff, the data would have been relative save if one of the mirror drives failed due to hardware fault.
Yeah, but that's a reliability thing - not a backup. A backup should allow you to recover data if you accidentaly type 'rm -rf *' in the wrong directory - which RAID will most certainly not cover.
Quote:
Originally Posted by [USRF]Obiwan
If there was for example a DLT tape backup. The same person could have destroyed all the DLT backup tapes along the deletion of files on the hard drives.
That would be why you have off-site storage of backups - again, in the event of a fire or flood your data is gone unless you've got a full copy somewhere else. It's not like you have to spend any money, either: just take a weekly tape backup home with you every Friday.
Quote liratheal 7th January 2009, 12:38
All I can say is: Noobs.
Quote Bluephoenix 7th January 2009, 12:55
Quote:
Originally Posted by liratheal
All I can say is: Noobs.

+1
Quote chicorasia 7th January 2009, 12:57
someone once said:

"Good judgement comes from experience.
Experience comes from bad judgement."
Quote phuzz 7th January 2009, 12:59
As an IT admin this story makes me cry.
And then double check all my backups.
Our SQL servers make a backup to disk every night, also the entire server is backed up to tape incrementaly every night, and then the tapes are stored off site, oh, and they all run on RAID5 volumes as well of course, that's almost what I'd consider a minimum for a business system.
Quote Venares 7th January 2009, 13:05
/FACEPALM

RAID (Redundant Array of Independent Disks)

Key word there being "Redundant ".
Its whole point is to prevent data loss due to a poped drive, not as a backup solution.
Any company that doesnt have a good Disaster Recovery plan in place and tested regularly is just ASKING for it.
Sad, bit I have to aggree with liratheal & Bluephoenix.
n00bs of the highest order.
Quote raGe82 7th January 2009, 19:34
At first I thought it's some joke... I just can't believe it - hosting company is 6 years in the business and no one ever at least read about storage and backups?
Quote LordPyrinc 7th January 2009, 23:45
Makes me wonder where they got there Sys Admins from... highschool career day?
Quote DXR_13KE 8th January 2009, 00:50
they went the way of the dodo... natural selection works....
Log in

You are not logged in, please login with your forum account below. If you don't already have an account please register to start contributing.