AAPL | Bach Seat

Tag Archive for AAPL

RB | October 4, 2012

Microsoft Eliminating Backup Generators

The venerable diesel backup generator has long been a symbol of reliability for mission-critical installations. Backup generators provide the emergency power required to keep servers online during utility power outages. Data Center Knowledge reports that the growing focus on using clean energy to power large data centers is prompting Microsoft (MSFT) and other tech titans to ditch their generators, along with their diesel fuel emissions.

Microsoft is the latest company to announce its intention to cut its use of diesel generators. The move is part of a broader initiative to make Microsoft’s server farms more sustainable and less reliant on the utility grid according to DCK. Microsoft Utility Architect Brian Janous wrote in a recent blog post.

We are currently exploring alternative backup energy options that would allow us to provide emergency power without the need for diesel generators, which in some cases will mean transitioning to cleaner-burning natural gas and in other cases, eliminating the need for backup generation altogether.

DKC speculates that the reference to natural gas suggests that Microsoft is preparing to add fuel cells to replace its generators. That could be good news for Bloom Energy, which recently scored wins to replace generators and UPS units at new data centers of eBay (EBAY) in Utah and supplement power Apple‘s (AAPL) data center in North Carolina.

DCK explains the Bloom Energy Server is a solid oxide fuel cell technology that converts fuel to electricity through an electrochemical reaction, without any combustion. The Bloom box can continue to run during grid outages because they are housed at the customer premises. Bloom boxes can run on natural gas or a range of other biofuels, including methane gas from landfills.

Another reason MSFT may be replacing generators according to DCK is that they have caused Redmond several headaches in recent years, including an Azure cloud outage in Europe (when multiple generators didn’t start during a utility outage) and public controversy about whether the diesel emissions from Microsoft’s generators in Quincy, WA could cause health problems for local residents. Diesel engine exhaust is a regulated pollutant and can be toxic in high concentrations.

Or Microsoft’s motivation could be to become less dependent on the utility grid and use renewable energy to power its servers the blog says. The company says its “data plants” will break new ground in integrating electricity and computing, bring together data centers and renewable power generation.

One type of renewable energy Microsoft has explored is a waste-powered data center. It could be built on the site of a water treatment plant or landfill. In his blog post, Janous indicated that Microsoft is evaluating a biomass project in Europe (rb- I wrote HP’s plan to use manure to run a data center here).

“Given the unreliability of the electric grid and the need for continuous availability of cloud services, Microsoft maintains diesel generator backup at all of our data centers…” Janous wrote. “These generators are inefficient and costly to operate. From both an environmental and a cost standpoint, it makes no sense to run our generators more than we absolutely must.”

Microsoft is also considering “long-term purchases from larger grid-connected installations that would displace some portion of our grid purchases,” Janous wrote. Google (GOOG) has embraced a similar strategy, using power purchasing agreements to add more than 200 megawatts of wind power to the local utility grids that support its data centers.

Microsoft is taking steps to position itself to make bulk power deals according to DCK. “We have recently signed on as an advisory board member with Altenex, an operator of a network that enables member companies to more efficiently engage with developers of renewable energy projects,” Janous said. “We expect this engagement with Altenex to improve our ability to identify and evaluate cost-effective clean energy projects.”

rb-

I recall as a newbie techie the first time I had to be in on Sunday morning to work with the site engineer to crank up the 100 HP Cummins standby generator. The firm ran the monthly test to make sure the critical systems stayed up. The generator was enclosed in a secure room that contained the heat and noise. The exhaust was vented out. One of my regular jobs was to kick the standby 55-gallon drum of diesel with the hand pump on it to make sure there was fuel available for the generator.

Google powers a data center with wind for the first time (gigaom.com)

Ralph Bach has been in IT long enough to know better and has blogged from his Bach Seat about IT, careers, and anything else that catches his attention since 2005. You can follow him on LinkedIn, Facebook, and Twitter. Email the Bach Seat here.

Category: Uncategorized | Tags: 2012, AAPL, Apple, Bloom Energy Server, Business Continuity, Data center, Disaster recovery, EBay, Green, Microsoft, MSFT

RB | September 22, 2012

Comments Off

SmartPhone Zombie Apocalypse

If you have a smartphone, online criminals may soon have your number. Smartphone malware is getting increasingly sophisticated, and MIT‘s Technology Review reports that a security researcher has created software that turns a smartphone into a “zombie” that can be controlled remotely. The blog says Georgia Weidman created the program, which controls a Google (GOOG) Android phone via short message service (SMS) to bring about a smartphone zombie apocalypse.

Once only theoretical, real-world cell-phone viruses are becoming more common. The article reports the most famous was a scam in Russia that tricked users into installing malicious software on Android phones and using the SMS functionality to send messages to a number that charged a premium fee. In late 2010, a Chinese virus for Android devices stole personal data according to the article.

Botnets have become a staple of Internet crime. They can be used to attack other systems, host attack tools, send spam, or just steal data. The blog says this type of attack has been rare with mobile devices, but that seems to be changing. “We have been taking down Internet botnets for years now, but there is not as much understanding [of telecom networking],” Ms. Weidman says. “I definitely see criminals going more and more toward using the telco’s network.”

TR explains that Ms. Weidman’s program is one of the first known to turn smartphones into zombie nodes of a botnet. Her attack works like this: After infecting a phone with a rootkit, she uses that phone to send spam text messages, takes part in a denial-o f-service, or degrade the communications of the phone—all without the user knowing. The techniques apply to any smartphone, Weidman says.

Today’s smartphones have multiple layers of defense. For one, they can block malicious applications. They also have managed channels, such as the Apple (AAPL) App Store and Google’s Android Marketplace, for applications.

As a result, Weidman says, infecting a smartphone is not easy. “The hurdle with any malware is infecting the phone,” she told Technology Review, noting that the methods used by cybercriminals usually do not work. “More of what you see of malware is people downloading applications for their phone that are infected,” she says.

The problem of cyber-criminals targeting consumers’ phones will only get worse Kevin Mahaffey, chief technology officer of mobile-security startup Lookout told the author. Because the control of phones is so easy to turn into cash via premium text messages, criminals will be drawn to attack the devices.

“I always tend to look at the economics of the problem to ask myself whether it will continue in the future,” the CTO explained. “And because there is an incentive for attackers to compromise mobile phones, and the cost of compromising is not that high, that says it will become more prevalent in the future.”

Using the telecommunications network, rather than the Internet, for botnet control allows attackers to hide their actions from users. When the attacker does it using malicious software, the user has little chance of detecting it, says Weidman.

“When I infected a phone in my botnet—my lab botnet—with malware, the smartphone would receive a message through SMS and I would check to see if it has botnet instructions in it,” she says. “If it does, it would perform the functionality requests, and then it would swallow the message, so the user does not know that there was a message at all.”

While phones do not have the computing power of more traditional computers, they are hefty enough to handle many of the tasks that cyber-criminals desire, she says. She adds that the sheer number of smartphones means that any botnet could be “a real threat” to create a smartphone zombie apocalypse.

Android Malware Infections Increase By 700% (escapistmagazine.com)

Category: Uncategorized | Tags: 2012, AAPL, Android, Apple, Botnet, GOOG, Google, Malware, Phone, Security, Smart, SMS, SPAM

RB | August 28, 2012

One comment

A History of Mac Malware: Part 1

Graham Cluley at Sophos recently wrote an excellent history of Apple Macintosh malware. He points out that Mac malware is a subject that raises strong emotions. There are some who believe that the problem is over-hyped and others who believe that the malware problem on Macs is underestimated by the Apple-loving community. The author writes that hopefully, this short history will go some way to present the facts and encourage sensible debate. (rb- We have just taken on a new customer which is 85% Mac and 15% PC. I have had this very conversation with my Apple certified tech who does the field support.)

Click here for part two of this series. Click here to read my recent series commemorating the 25th anniversary of the computer virus.

1982 – Apple II – The first virus to affect Apple computers wasn’t written for the Macintosh (the original Mac did appear until 1984). 15-year-old student Rich Skrenta wrote the Elk Cloner virus, capable of infecting the boot sector of Apple II computers. On every 50th boot the Elk Cloner virus would display a short poem:

It will get on all your disks
It will infiltrate your chips
Yes, it’s Cloner!

It will stick to you like glue
It will modify RAM too
Send in the Cloner!

The blog says many Apple fans are surprised that the Elk Cloner boot sector virus predates IBM (IBM) PC viruses by some years. (I got my first paying tech job using an Apple II and PFS:File to build a database).

1987 – Macintosh – The nVIR virus began to infect Apple Macintosh computers, spreading its malware mainly by floppy disk. It was a similar story to what was happening in the world of MS-DOS malware, where viruses would typically travel from computer to computer by users sharing floppy disks.

Source code for nVIR was later made available, causing a rash of variants for the Mac platform. The author writes that the first anti-virus products for Mac, some free, some commercial, began to emerge in response th this malware. (In my first tech support Job, I got very familiar with the Mac 30/SE, since there was a computer lab full of them with a SCSI chain from the Mac to an external hard drive to a scanner. They also printed to a LaserWriter 2 with AppleTalk and Phonenet. I still have a bag of terminators.)

1988 – HyperCard – Running on early versions of Apple’s Mac OS, one HyperCard virus displayed a message about Michael Dukakis’s US presidential bid before self-destructing:

“Greetings from the HyperAvenger! I am the first HyperCard virus ever. I was created by a mischievous 14-year-old, and am completely harmless. Dukakis for preseident (sic) in ’88. Peace on earth and have a nice day”

1990 – The MDEF virus (aka Garfield) emerged, spreading malware on application and system files on the Mac.

1991 – HC (also known as Two Tunes or Three Tunes) was a HyperCard virus discovered in Holland and Belgium in March 1991. The writes that on German language versions of the operating system it would play German folk tunes and display messages such as “Hey, what are you doing?” and “Don’t panic.”

1995 – Concept Macro Virus – Microsoft (MSFT) accidentally shipped the first-ever Word macro virus, Concept, on CD-ROM. It infected both Macs and PCs running Microsoft Word. Concept was not written with malicious intent but thousands of macro viruses were to follow, many also affecting Microsoft Office for Mac. Word macro viruses turned the world of Mac *and* Windows malware on its head overnight according to Sophos.

Macro viruses are written in an easy-to-understand macro language that Microsoft included in its Office programs making it. The blog says the macro language made it child’s play to create new malware variants. Most people at the time considered documents to be non-dangerous and were happy to receive them without thinking about the security risks. Just opening a Word .DOC file could infect your computer because the macro virus’s code was embedded within.

1996 – Laroux Excel macro virus – The Laroux virus did not affect Mac users until Microsoft released Excel 98 for Mac and then Apple users could also become victims.

1998 – Hong Kong introduced the next significant Mac malware outbreak the blog says. It was first spotted in the wild in Hong Kong. The worm – dubbed AutoStart 9805 – spread rapidly in the desktop publishing community via removable media, using the CD-ROM AutoPlay feature of QuickTime 2.5+. (rb- An AutoPlay issue – whoda thunkit?). In the same year, Sevendust, also known as 666, infected applications on Apple Mac computers.

After 1988 Mr. Cluely writes that big changes to the Mac malware scene were just around the corner. The release of Mac OS X, a whole new operating system which would mean that much of the old malware would no longer be capable of running. Mac-specific malware would have to be written with a new OS in mind.

In Depth: Macs and malware: how real is the threat? (techradar.com)

Category: Uncategorized | Tags: 2012, AAPL, Apple, Elk Cloner, history, HyperCard, IBM, Mac, Mac OS, Mac OS X, Macintosh, Malware, Microsoft, MSFT, Security

RB | August 23, 2012

One comment

Privacy on IPv6 Networks

Internet service providers, websites, and equipment vendors around the globe took part in the World IPv6 launch in June, Internet companies including AT&T (T), Cisco (CSCO), Comcast (CMCSA), Facebook (FB), Google (GOOG), Microsoft (MSFT), Verizon Wireless (VZ), and Yahoo (YHOO) decided to permanently turn on IPv6. A small fraction of Internet users and devices have started communicating via IPv6 networks, with more and more transitioning to the new protocol over the coming months and years. There are security and privacy implications in the switch to IPv6.

All kinds of devices will get new IPv6 numbers as the addressing format grows. The IPv6 addresses for these networked devices can be generated in a number of different ways and the choice of how they are created has potentially wide-reaching effects for security and privacy Center for Democracy & Technology explains. One of the original methods for assigning new addresses involved using a unique device identifier (known as a MAC address) as the suffix of the IPv6 address. This method creates a permanent, unique address for a device, potentially allowing any server that the device communicates with to indefinitely track the user.

IPv6 designers soon realized the potential security and privacy problems of MAC-based addresses; as a result, they created an alternate method known as “privacy extensions” or “privacy addresses” the article reports. The privacy extensions use a randomly generated number instead of a MAC address. In order to protect privacy on an IPv6 network, the random number is unrelated to any device identifier and in practice lasts no more than a week (and often much less time), ensuring that the user’s IP address cannot be used for long-term user tracking.

It is up to operating system vendors to choose which IP address assignment method will be the default on their devices. The author says that some vendors have made good choices, particularly within the last year. Microsoft has long led the charge on IPv6 privacy, with privacy extensions on by default in all versions of Microsoft Windows since the release of Windows XP nearly a decade ago. Apple followed suit last year, with privacy extensions activated by default in all versions of Mac OS X since 10.7 (Lion) and with the release of iOS 4.3 for iPhone and iPad. Google did likewise in its Android 4.0 release last year.

The CDT says that as long as Internet users choose to upgrade their operating systems to the latest versions, they should be protected against perpetual security and privacy threats from IPv6 network address tracking.

rb-

However, I wrote about reports from H.Security that mobile operating systems do not protect security or privacy on IPv6 networks. The report says mobile OSs send private information about their users to the network. The H.Security article says this is not a flaw in IPv6, rather it is lazy programming in some cases. The article points out that neither Apple’s iOS nor Android devices have the option to enable Privacy Extensions or the option to disable IPv6. apparently, the only thing smartphones need is a control option in the user interface to protect mobile OS users’ privacy and security on an IPv6 network.

Romania Has the Fastest IPv6 Adoption Rate (maindevice.com)

Category: Uncategorized | Tags: 2012, AAPL, Apple, AT&T, Facebook, FB, GOOG, Google, IPv6, Microsoft, MSFT, Privacy, T, Windows, XP

RB | July 17, 2012

3 comments

IPv6 Compromise Smartphones Users’ Privacy

IPv6 Compromise Smartphones Users' Privacy Now that the IPv4 address pool is depleted and the IPocalypse is at hand, wrinkles are emerging in IPv6. One of the wrinkles is with mobile devices. Most of the cool mobiles devices have been able to handle IPv6 for a while. Apple’s (AAPL) iPhones, iPads, and iPods have been capable of handling IPv6 Since version 4 of the iOS operating system and most Google (GOOG) Android devices have been capable since version 2.1. H Security is reporting that these mobile operating systems send information about their users to the network.

A device on an IPv6 network usually determines half of their address (the “interface identifier”) themselves, but H Security says that smartphones are sloppy with this task. According to the article, smartphones simply add the same two bytes to their globally unique MAC address and use it as their identifier. As a result, they transfer a unique hardware ID whenever they communicate with an IPv6-enabled server.

The basic problem isn’t an IPv6 issue because there are other methods for generating the address. The article says that a device can generate a random interface identifier and replace it on a regular basis. This is called the Privacy Extensions method and is the factory-set option in Windows; it can also be enabled in other operating systems. The article points out that devices running Apple’s iOS or Android offer neither the option to enable Privacy Extensions nor the option to disable IPv6, anyone who uses an affected device on an IPv6-enabled wireless network will send their ID.

The only thing the smartphones are lacking is a control option in the user interface, as the Privacy Extensions do come as part of their kernel. For instance, on a (jailbroken) iOS 4 device with root access, they can be enabled with the same command that enables them on a desktop device running Mac OS X:

sysctl -w net.inet6.ip6.use_tempaddr=1

The blog claims the problem is only affecting a small number of users because IPv6 is not yet in widespread use. However, more ISPs plan to offer IPv6 in addition to the old IPv4 in the future. In addition, there are routers like the Cisco (CSCO) Linksys E3000, which will automatically set up an IPv6 connection via a 6to4 conversion when their internet access is purely IPv4.

The author concludes that the issue is particularly sensitive because such devices tend to be used by one specific person. As a result, the MAC address, which is accessible to any server operator and network monitor, allows this user to be identified.

rb-

If this sounds familiar, it is I wrote about mobile apps uploading UDID’s here.

IPv6 and click fraud (net-security.org)
Windows 8 moves to IPv6 Internet (zdnet.com)

Category: Uncategorized | Tags: AAPL, Android, Apple, GOOG, Google, iPad, iPhone, IPocalypse, iPod, IPv6, Networking

« Older Entries Recent Entries »

Bach Seat

Tag Archive for AAPL

Microsoft Eliminating Backup Generators

Related articles

SmartPhone Zombie Apocalypse

Related articles

A History of Mac Malware: Part 1

Related articles

Privacy on IPv6 Networks

Related articles

IPv6 Compromise Smartphones Users’ Privacy

Meta