Threat Lab

Vaporizers (AKA E-cigarettes) have been gaining some serious traction and widespread use over the past few years. The sudden surge of popularity isn’t too surprising considering the fact that the health implications of nicotine consumption are vastly more favorable with vaporizers when compared to traditional cigarettes.

Most Vaporizers charge through a propriety connection to USB that looks something like this:

Should be harmless, right?

In a recent reddit post, the poster reported that an executive at a large corporation had a data security breach on his system from malware, the source of which could not be determined initially. The machine was patched up to date, had updated anti-virus protection, and Weblogs were evaluated. “Finally after all traditional means of infection were covered; IT started looking into other possibilities…” The made in china USB charger had malware on it that, when plugged into a computer’s USB port, would phone home and infect the system.

Now for those of you scratching your head going – hang on a minute… Windows hasn’t auto-executed anything from USB in YEARS. USB drivers are loaded from the library on the PC and I would know when it was plugged in and I would have to click and run a file in that folder – this whole story sounds fishy… Let me introduce you to BadUSB. Essentially this USB control chip would be reprogrammed to act as a keyboard + mass storage device. Once plugged in, it sends key-commands to open command prompt and then executes files from the storage. It’s not as if this vector of attack is brand new either – at least conceptually. According to @th3j35t3r (the Jester), a well known cyberwarrior in an article titled ‘What would I do if I was Chinese PLA’, USB charger attacks such as this are “theoretical but entirely possible, if not probable”.

My personal suggestion to those concerned is to only charge USB devices through a wall adapter (they charge faster anyway). If you REALLY need to charge through USB then I suggest getting one of these, dubbed “USB Condoms”, which will make sure that only power is drawn and no data is exchanged.

What kind of defenses exist for this type of attack? Basically not much. Malware scanners cannot access the firmware running on USB devices and USB firewalls that block certain devices do not exist yet. Behavioral detection is unlikely since the device’s behavior is just going to appear as though a user has simply plugged in a new device. It’s very unsettling and the threat is there however unlikely we think it is. While I doubt this is widespread or even remotely common, I did make sure to take apart my charger and made sure that there were no data pins and that it was only drawing power through USB.

The holiday season is almost upon us, which means the holiday shopping season is also almost upon us.

And as always, it’s bound to be a crazy time of scrambling for the biggest and best deals, both in stores and online.

But while your wallet is destined to take a hit as you stack up on gifts for your family and friends, you want to make sure cybercrooks don’t make your list of people who will be receiving presents this year.

Sadly, with 2014 being labeled by some as ‘The Year of the Hack’, it may be easier for them than ever before to do just that. Fortunately, armed with some general security know-how, you can make their hacking jobs significantly harder while making your online shopping experience slightly less stressful.

Here are 5 online shopping tips to stay secure this holiday season:

Be Breachophobic

This one applies to traditional holiday shopping as well…

With the influx of massive data breaches across a wide variety of industries, no company seems to be safe. And popular retailers have been hit particularly hard (looking at you, Target, Michaels, Home Depot, Staples, KMart, etc).

Unfortunately, these breaches don’t show any signs of slowing down. Perhaps even more unfortunately, as a result of this, consumers are experiencing ‘breach fatigue’ and not changing their buying behavior even in the midst of all these attacks (according to a recent report from Ponemon).

But in this case, fear is actually a good thing. It keeps you on your security toes. Don’t have the ‘yeah, a lot of people are impacted, but it won’t happen to me’ attitude when it comes to breaches.

If you learn of a breach at a company whose store you recently bought something in or at a bank that you use, take a proactive approach. Call you credit card company and have a new card issued ASAP. Call your bank and find our what steps you need to take to protect yourself. These are not the most fun activities, but they could save you from a potential financial/data loss nightmare.

Likewise, if you’re planning to go shopping at a particular retailer and you find out they’ve recently experienced a data breach, look to do your shopping elsewhere. Ok, maybe you won’t be able to take advantage of that exclusive Black Friday deal, but most of the competing big-name stores will likely have something similar. Plus, what good is that brand-new big-screen TV you got for a ‘steal’ if attackers got access to your credit card number?

Beware of enticing ‘Amazing Deal’ links in your email inbox

If a deal looks too good to be true, it probably is.

Phishing emails are still a popular tool for cybercriminals. The difference these days, however, is that they look more legitimate than they did in the past. Obvious red-flags like blatant grammatical errors or strange email addresses from the sender are less common and the bad guys are finding more creative ways to get you onto their fake, information-stealing pages.

An example of a UPS Phishing email (Source: PC Mag)

And the fact that the holiday season has many legitimate great deals actually benefits cybercriminals, giving them a perfect opportunity to blend in with a phishing email that would normally seem out of place.

So don’t just immediately click a link in that ‘great deal’ email that popped up in your inbox. Verify that the sender is legitimate, check for grammatical errors and inconsistencies, and even compare it with another email you’ve gotten from that company in the past that you know was legitimate. If there’s something odd or out-of-place, don’t pull the purchase trigger and potentially open yourself up for identity theft or money loss.

Know your surroundings

Just because you can connect to WiFi almost anywhere doesn’t mean you should connect to WiFi any time it’s available.

If you’re doing any of your holiday shopping online, it’s really in your best wishes to do it over a secure network.

As appealing as it may seem to knock out some of your holiday shopping as you’re sitting sipping your mocha at that coffee shop, know that public WiFi hotspots are just that – public. That means anybody can connect to them, including an attacker looking to catch a hapless holiday shopper unawares.

There’s a much lower chance someone will break into your network at home, unless of course your home network isn’t password protected. Which brings up another good tip – password-protect your WiFi network at home. And make sure to actually use a strong password! No ‘password’ for your password.

Mobile Devices are vulnerable, too

PC, Mac, Android, iOS…it doesn’t matter. No device/operating system is malware or hacker-proof. Let me rephrase: that means mobile devices aren’t in the clear! No, not even the iPhone is safe; remember the recent WireLurker malware?

So if you’re thinking: “I’ll just do my all my online holiday shopping on my tablet to avoid the chances of getting hacked”, don’t do that. It’s a flawed mindset.

Mobile is a hot target for hackers at the moment. Remember that today’s mobile devices aren’t the bricks of yesteryear that you used to make calls and play Snake. Nope, today’s smartphones (and especially tablets) are bonafide computing machines, and protecting them in the same way you’d protect your computer isn’t an idea that should be ignored.

But aside from installing mobile security (which you should definitely do), there are other actions you can take to mitigate mobile risks, a major one being not jailbreaking/rooting your devices and/or using third-party app stores. Easier to do and access on Android devices, these third-party app stores are often riddled with malicious apps that can steal your information and dollars.

Safeguard all your devices

Even if you’re intelligent in your browsing and downloading habits, having computer ‘street smarts’ isn’t enough anymore. New threats are emerging seemingly by the hour, vulnerabilities like HeartBleed and Shellshock are coming into light, and if you fall victim to an encrypting ransomware (that seems to be ever-evolving) with no sort of protection, you’re paying hundreds of dollars to get your files back or paying even more for a new computer.

But let’s bring it back to the holiday shopping topic. Take the fake phishing email scenario, for example. Even if you take the aforementioned steps to verify the validity of the email, you’re not completely out of the ‘cyber’ woods. Like I said, hackers are becoming increasingly more clever and those phishing emails are often very difficult (if not flat-out impossible) to distinguish from the real thing. You need an intelligent security solution to have your back in case you get tricked despite your best efforts.

The holiday shopping season is less than a week away so be sure that you’re ready and that your devices are secure. Follow these basic online shopping security tips and go out there and buy those presents…carefully, yet confidently! And by ‘go out there’, I mean on the internet. That way, you can avoid this:

(Source: CNN Money)

Happy holiday shopping!

CoinVault

Today we encountered a new type of encrypting ransomware that looks to be of the cryptographic locker family. It employs the same method of encryption and has a very similar GUI (kills VSS, increases required payment every 24hr, uses bitcoin payment, ect.).

CoinVault GUI

Here is the background that it creates – also very similar.

What’s unique about this variant that I wanted to share with you all is that this is the first Encrypting Ransomware that I’ve seen which actually gives you a free decrypt. It will let you pick any single file that you need after encryption and will decrypt it for you.

It even works too! How nice of of them.

This is a really interesting feature and it gives a good insight into what the actual decryption routine is like if you find yourself actually having to pay them. I suspect that this freebie will increase the number of people who will pay.

Webroot will catch this specific variant in real time and heuristically before any encryption takes place. We’re always on the look out for more, but just in case of new zero day variants, remember that with encrypting ransomware the best protection is going to be a good backup solution. This can be either through the cloud or offline external storage. Keeping it up to date is key so as not to lose productivity. Webroot has backup features built into our product that allow you to have directories constantly synced to the cloud. If you were to get infected by a zero day variant of encrypting ransomware you can just restore your files back as we save a snapshot history for each of your files up to ten previous copies.

There is a clear trend that every year there are bigger targets compromised with increased frequency with more personal data being collected. The recent attack on JP Morgan Chase is especially alarming considering they are the largest US bank and hackers had gained access to numerous servers with administrative access for nearly a month before being detected.

According to reports, JP Morgan Chase account details for every consumer and business customer, including to name, address, email address and phone number, were compromised. Fortunately more specific details such as account numbers, social security numbers (SSNs) and passwords were not accessed. This is due to much stronger security measures being implemented to keep this data safe.

While the nature of the breach have not been disclosed, the company may have been targeted by a group of Russian hackers. Often these attacks are very well planned, targeting a few key employees for the first stage of the breach. These individuals are analyzed to understand their web habits.

Based on their analysis, an attack will be staged to either infect a target system or to gain credentials. Often it happens by attacking a website that is frequently visited by the target as security of an individual website is far less than that of a big corporate bank.

To reduce the impact of this attack, JP Morgan Chase could have applied the same level of encryption and security to user account details that they applied to account numbers, SSN’s and passwords. Another option would be to enforce a strict two factor authentication policy for all network administrators.

Overall, there is a clear trend of attacks that aim to compromise point of sale systems as well as companies who store vast amounts of user data. These businesses need to prepare for continued attacks by updating their security policies and systems to be on high alert.

Recently during some research on encrypting ransomware we came across a new variant that brings some new features to the table. It will encrypt by utilizing the following javascript from being opened as an attachment from email (posing as some document file).

Malicious script from email

Once full encrypted you’ll get a popup text document informing you that all your files have been encrypted and how to pay money to get your key to decrypt. This specific sample is Russian, and the instructions were also in Russian so I didn’t show it here. The really interesting thing about this variant that I wanted to share is that once it finishes it actually shows you a twitter feed that populates a tweet every time someone pays the ransom. I suspect this does increase the chance that people will pay the ransom.

Blurred out emails of the victims who paid

Webroot will catch this specific variant in real time before any encryption takes place. We’re always on the look out for more, but just in case of new zero day variants – remember that with encrypting ransomware the best protection is going to be a good backup solution. This can be either through the cloud or offline external storage.  Keeping it up to date is key so as not to lose productivity. Webroot has backup features built into our product that allow you to have directories constantly synced to the cloud. If you were to get infected by a zero day variant of encrypting ransomware you can just restore your files back as we save a snapshot history for each of your files up to ten previous copies.

Everyone’s heard the saying ‘Macs aren’t malware-proof’, right? Oh, you haven’t? Count me not surprised.

It could be due to the fact that that’s not an actual saying, but the more likely reason is that there is a deep-rooted belief among Apple users that Apple products and services are somehow, by default, impervious to viruses, malware, hackers, etc. Allow me to reiterate: THEY’RE NOT!

Need proof? Well, we could look back two years ago to Dexter (not the hit Showtime show). Or let’s rewind to early this year, when researchers learned that Macs were still vulnerable to 2011’s famous ‘Flashback Trojan’? Or how about iDroid, the mobile trojan capable of infecting both Android and iOS devices? Then there was September’s iCloud hack fiasco that started the wave of nude celebrity photos across the world wide web. And then, just yesterday, it came to light that two of Apple’s security tools, Gatekeeper and XProtect may not pick up the recently-discovered iWorm malware.

But just in case all that isn’t enough, enter WireLurker, the newest piece of evidence to back up the ‘Mac-and-iOS-malware-is-real’ school of thought.

(Source: Mashable)

Discovered by security researchers from Palo Alto Networks (you can download the report here), WireLurker is a new family of malware that was found in the wild on a popular third-party Chinese app store called Maiyadi, and that allows hackers to infiltrate iOS devices through an infected Mac via a USB connection (only the second known instance of this kind of iOS attack). But what truly makes it unique is that it’s able to infect iOS apps in a fashion similar to a traditional virus and, even more importantly, it doesn’t require for the iOS device to be jailbroken to do its damage.

According to the report, WireLurker has already ‘trojanized’ 467 OSX apps in the Maiyadi App Store. Those apps have been downloaded 356,104 times.

The (relatively) good news is that this malware attack is pretty limited in scope. We were able to secure a quote from Ryan Olson of Palo Alto Networks, one of Webroot’s technology partners:

“Our lead researcher on this effort, Claud Xiao, first began investigating the malware after seeing reports of strange activity on Mac OS and iOS devices in China. He tracked the attack back to it’s source and took a deep dive into the malware to discover how it works. We don’t expect this WireLurker malware to spread to the United States, but the tactics that it uses are likely to be copied by new attackers who could have new targets in mind.”

An Apple spokesperson offered some more good news:

“We are aware of malicious software available from a download site aimed at users in China, and we’ve blocked the identified apps to prevent them from launching. As always, we recommend that users download and install software from trusted sources.”

The fact that WireLurker is only targeting users in China may be a sign of relief for the West, but, as Olson pointed out, it doesn’t mean a similar attack couldn’t be expanded to wreak havoc outside of China. iOS (unlike OSX) has had a fairly spotless record when it comes to malware up until this point (if you don’t factor in ‘jailbroken’ devices), but WireLurker is the first real virus scare for the OS because it doesn’t require a jailbreak to infect the device, making it an exception to the ‘the Apple App Store is the best protection against viruses’ rule.

So what does all of this mean for Apple device users?

Michael Sweeting, one of our Senior Threat Analysts and who specializes in Apple devices sums it up nicely:

“End users of non-jailbroken iOS devices are typically unable to load third party applications, and are therefore somewhat protected from malware attacks propagated through app installs. OSX.WireLurker however is a new approach to infecting iOS devices with information stealing malware, by infecting third party application installers for OSX, and in turn infecting the iOS device once connected to the Mac via USB. This type of infection should once again remind Mac and iOS users alike that these devices are not immune to attacks, and even bigger more sophisticated attacks may be on the horizon. Users should continue to exercise caution and prudence, and should avoid downloading Mac software from third party app stores.​”

That ‘caution and prudence’ doesn’t only mean updating iOS, avoiding connecting your iPhone or iPad to unknown/untrusted computers, and not jailbreaking your device and sticking to the App Store (all of which are encouraged security tips, by the way). It may also mean safeguarding your Mac with an internet security solution.

Or maybe most importantly, it may mean finally accepting that Apple devices aren’t unhackable machines. But hey, if you already exercise the former precautions, you probably already take Apple security seriously and the last point likely doesn’t apply to you.

Just when consumers were starting to regain some company trust and safe-shopping stability after last year’s massive Target breach, a string of new large-scale company breaches quickly reminded us consumers just how insecure our personal data can be.

Needless to say, it’s been a rough year for some major companies and an even rougher year for thousands of unlucky customers. Let’s look at three of the major breaches of the last couple of months.

Home Depot

(Source: Krebs On Security)

Early last month, reports started coming in that the home improvement giant was investigating “some unusual activity with regards to its customer data.” Security reporter Brian Krebs immediately called credit card breach, especially since multiple banks came out to say that they were seeing evidence that Home Depot was the likely source of a batch of stolen credit and debit cards that went on sale in the cybercrime black market that morning.

Sure enough, six days later, the company admitted that its payment systems were in fact breached and that the hack was going on for months. They went on to say that while credit card data was exposed, personal pins were not. Reassurance (not really). And while the exact number of affected cards wasn’t known at that time, one thing was for certain: If you used a credit card at one of Home Depot’s U.S or Canadian stores in the past 4-5 months, you needed to consider your credit card stolen and get on the phone with your bank ASAP.

About two weeks later (September 18th), Home Depot announced the number. A whopping 56 million cards were impacted, making the incident the biggest retail card breach…ever (on record, at least). The ‘silver lining’? Home Depot also said that the malware was now contained.

Japan Airlines

(Source: Google Images)

Before the month of September passed (and with Home Depot still fresh on everyone’s minds), another large company from a completely different industry had some bad news to share with its customers…

On September 30th, Japan Airlines (JAL) confirmed that as many as 750,000 JAL Mileage Bank (JMB) frequent flyer club members’ personal info was at risk thanks to a breach. Apparently, hackers were able to get into JAL’s ‘Customer Information Management System’ by installing malware onto computers that had access to the system. The data that was accessed? Everything from names to home addresses to birthdates to JMB member numbers and enrollment dates. The good news is that credit card numbers and passwords did not appear to be exposed.

There have not been any new developments about this breach, but you here is the official statement by JAL from September 29th.

JP Morgan 

(Source: Reuters)

October 2014 was only two days young when yet another major company confirmed a data breach. This time, the victim was JP Morgan. Or rather, JP Morgan customers who used Chase.com and JPMorganOnline websites, as well as the Chase and JP Morgan mobile apps.

Last Thursday, the nation’s largest bank revealed that a mid-August cyberattack exposed personal info for 76 million households, as well as 7 million small businesses. More specifically, names, email addresses, phone numbers and addresses were stolen, while JP Morgan went on to say that there was no evidence that account numbers, passwords, Social Security numbers or birthdates were exposed. While the bank found out about the breach of it’s servers in August, it has since been determined that it began as early as June.

Unfortunately, not much else is certain at this time. What we do know is that Russian hackers are suspected (still not confirmed), over 90 over JP Morgan’s servers were affected, and it is believed that nine other financial institutions were also targeted (although we don’t know their identities). The lack of concrete information is scary in it’s own right, but the fact that JP Morgan is staying mum on the matter is even more troubling. According to a Huffington Post report from earlier today, the bank is refusing to say how many people were actually hit by the breach, with spokeswoman Trish Wexler saying that JP Morgan isn’t offering more details beyond what was announced last Thursday. This could mean that the breach, already the largest (against a bank) in history, could potentially be even larger than the reported 76 million households and 7 million small businesses, keeping in mind that ‘households’ is not the same thing as ‘individuals’.

Additionally, Fox Business is reporting that the bank is now bracing for a massive-scale spear-phishing campaign in the wake of the breach. according to their sources. Considering that no bank info was compromised in the original breach (JP Morgan said in a statement that they haven’t “seen unusual fraud activity related to the incident”), this is a plausible next-step. Using the personal info obtained in the ‘first wave’, the attackers can send out legitimate-looking emails to the affected customers that say there is a problem with the user’s account and ask for Social Security numbers, passwords, etc. Alternatively, the emails could ask the customer to click an embedded link to update their account info, but in reality, the customer is taken to a official-looking fake site from which the attackers can nab the important financial information. In either case, the virtual trap is activated at that point.

What to do?

It’s no secret that data breaches are on a steep rise. According to a the Identity Theft Research center, there have been 579 data breaches this year, 27.5% more than there were at this time last year. And that number is only going to continue to increase.

In any of these three breaches, it’s important for customers to take basic security steps to ensure their information is safe, whether that means calling your bank and getting a new credit card issued (in the case of Home Depot), changing your password if you’re a JAL frequent flyer and JMB club member, or changing your log-in information and monitoring your online accounts if you bank with JP Morgan or Chase.

As more and more people choose to bank online (and become more internet-dependent in general), it’s also no secret that employing powerful and always up-to-date internet security on your devices is more crucial than ever before. Company breaches and spear-phishing attacks aren’t going anywhere. Take the necessary steps to keep your personal info protected!

Update: Apple has patched the Bash bug. For more info (including links to download the updates on your Mac), check out this TechCrunch report.

As of last week, there’s a new security bug in the news, and it’s wreaking havoc on the Internet.

(Source: Macworld/Errata Security)

Discovered by Stephane Chazelas, a security researcher for Akamai (who revealed the bad news to the world last Thursday), the ‘Bash bug, or ‘Shellshock’, is a particularly nasty vulnerability affecting the Bourne-Again Shell (thus the Bash acronym) of certain versions of the Unix and Linux operating systems. Yes, that includes derivatives like the Mac OS. In other words, it’s everywhere, and could affect a countless number of devices that connect to the internet.

Remember Heartbleed? This is scarier. And potentially a bigger deal, too.

Why? According to Robert Graham of Errata Security:

“Unlike Heartbleed, which only affected a specific version of OpenSSL, this bash bug has been around for a long, long time. That means there are lots of old devices on the network vulnerable to this bug. The number of systems needing to be patched, but which won’t be, is much larger than Heartbleed.”

Oh, and it IS old. How old? Graham said in a different blog post that the bash issue has been around for 20 years.

For the record, the National Vulnerability Database gave Bash a 10/10! Here’s what you should know.

What is it?

As previously mentioned, it’s vulnerability, a bug affecting the Bash shell used in many Unix and Linux operating systems. Think of the Bash shell as the command-line shell. This means that a hacker could take over and issue remote commands to web servers that aren’t patched. As a result, private information could quickly become public information.

Is every device running Linux or Unix vulnerable?

No. According to Rapid7 Global Security Strategist Trey Ford, there are certain requirements that make a server vulnerable. More specifically, servers capable of passing commands over the internet remotely, are susceptible.

What makes it so dangerous?

Bash may not affect as many devices as Heartbleed, but that may be the only consolation. Because (unlike Heartbleed) Bash lets hackers execute commands remotely, the repercussions could be a lot more serious.

Who should be worried? Should you?

While network administrators who manage internet presence for their companies (particularly those running a CGI app written in Bash or using Bash script), should be concerned, the everyday desktop user probably doesn’t have as much to worry about. That is unless you have a bunch of connected ‘Internet-of-things devices’

What about those Internet-of-Things Devices?

This is largely about patching updates, so while the aforementioned everyday desktop user may not be affected by Bash, that doesn’t necessarily mean his or her connected devices aren’t vulnerable. From home automation systems to routers to webcams to refrigerators, there are many Internet-of-things devices that use the Bash script. And most users would probably never think twice about installing software updates on something like a fridge. Now they should.

What’s being done? 

Companies like Google and Amazon were quick to react, rushing Thursday to patch this latest vulnerability on their end. And because recent versions of the Mac OS are vulnerable, Apple quickly responded, saying that while most OSX users aren’t at risk, the company was quickly working on patching the flaw. You can read more about that here.

What should you do?

Unfortunately, there’s not much you can do as this is a widespread vulnerability rather than say, a sneaky piece of malware. And because this is the case, you, as a consumer, can’t contain it by yourself. Rather, it’s up to the those maintaining the web servers. But absolutely do pay attention, run updates, and look for notifications from service providers who are vulnerable to Bash. And if a company tells you to take action and change your password because their servers are affected, listen to them and get on that immediately.

What about Webroot Servers?

We took all the necessary precautions and upgraded all of our systems to the patched version. We can verify that none of our services are susceptible. You can find our official statement on this support section of our site.

We’ll be keeping a close watch to see how this story unfolds, but in the meantime, take the necessary steps to ensure you’re as protected as possible and install necessary updates as they become available.

It seems as though every few weeks we see a new encrypting ransomware variant. It’s not surprising either since the business model of ransoming files for money is tried and true. Whether it’s important work documents, treasured wedding pictures, or complete discographies of your favorite artists, everyone has valuable data they don’t want taken.

This is the last thing anyone wants to see.

This variant does bring some new features to the scene, but also fails at other lessons learnt by previous variants. Starting with the new features this variant will now just “delete” the files after encrypting them (it just hides them from you). This doesn’t add any more intangibility since they are encrypted with AES-128 anyway, but it does add a greater sense of loss and panic since all of your common data directories will appear to have been cleaned out. Another new feature is the constant raise in price every 24 hours. While price bumping was used on previous variants, this one doesn’t have a limit and will increase by .2 bitcoins (~$97) every 24 hours until you crack or make peace with this loss.

Where this variant falls short on overall volatility is in the failure to delete the VSS (Volume Shadow Service) so using tools like Shadow Explorer will work to retrieve your files and circumvent paying the ransom. As I’ve said in previous blogs I do expect issues like this to be fixed once this malware is adopted by more botnets for widespread distribution.

Webroot will catch this specific variant in real time before any encryption takes place. We’re always on the look out for more, but just in case of new zero day variants – remember that with encrypting ransomware the best protection is going to be a good backup solution. This can be either through the cloud or offline external storage.  Keeping it up to date is key so as not to lose productivity. Webroot has backup features built into our product that allow you to have directories constantly synced to the cloud. If you were to get infected by a zero day variant of encrypting ransomware you can just restore your files back as we save a snapshot history for each of your files up to ten previous copies.