Security – Page 6 – Frank Leonhardt's Blog

Is HSBC’s voice identification really secure?

I was woken by Radio 4 this morning with news that HSBC (and First Direct) will be rolling out voice identification software as a replacement for the “cumbersome” password-based system currently in use. I’ve been using this cumbersome system for more than twenty years, and I can’t say I have any problem with it – ten seconds and you’re in; and time has proven it reasonably secure.

But this new biometric “voice-print” system sounds a tad more dodgy to me. It comes from Nuance Communications, and apparently it checks over 100 unique identifiers in someone’s voice, including speed and behavioural features and maps the sound it’s hearing back to physical features such as the shape of the larynx and nose. The technology might be better remembered as Dragon Dictate from the 1990’s, although Nuance has been working on the biometric aspects for some time, and recently announced Santander was going to use it in Mexico.

I’m naturally suspicious of any biometric identification method apart from retinae scans, having looked at many such schemes over the years. They’re generally vulnerable to amounts to “replay” attacks. Fingerprint or face recognition can usually be fooled relatively simply with a picture of the real thing. So what’s to stop a replay recording of someone’s voice? Nothing, as far as I can tell.

When the BBC asked about recordings being played back they were told that any recording process would lose the subtleties of live speech, and the BBC seemed happy with that. Well I’m not! The way telephones work these days, your voice is sampled, encoded in to very few bps and sent. How is this going to look any different to a recording? You can store and repeat a section of telephone call digital data easily enough and it’s bound to be indistinguishable.

I can see some solutions – the system could ask you to repeat some random phrase back instead, and word recognition could determine whether you said the right thing after the biometric recognition matched the voice print. But this isn’t the answer the BBC got.

I’m awaiting more information…

HSBC had a bad January with cyber-attacks. Is this some ill-conceived scheme to try and change the news agenda?

29-January-1631-January-16

Lincolnshire Council in £1M ransomware plot

Coat_of_arms_of_Lincolnshire_County_Council Reports are that Lincolnshire Council has been shut down for four days because it’s been targeted by ransomware that has encrypted all its files. That they’ve been a victim of such a scam doesn’t surprise me – it’s all too common. What’s moving my eyebrows skyward is the fact that the criminals are asking for £1M to restore their data.

I’ve seen a lot of this before, and the criminals generally ask for a sum that it’s easier to pay than mess around trying to repair the damage. In other words, £500 is normal but £1M is not. For this to be credible, someone would have had to target them specifically, and come up with a plot to damage a lot of data in one go. This is possible if one PC has R/W access to a lot of files on a server, but for the criminals to expect to do this value of damage the council would have to be pretty incompetent and the criminals would have had to know this for certain. (What am I saying?)

From the BBC report there are a couple of interesting lines:

“The authority said it was working with its computer security provider to apply a fix to its systems.”

Hmm. So who is their computer security provider? If they have one that’s any good, the network would have been set up to avoid such wholesale damage. Serco took over the Council’s IT operations in April 2015. in a £70M+ deal. Whether the outsource company has outsourced the “security provision” is a little harder to know.

Further down the BBC article it says:

“Chief information officer Judith Hetherington-Smith said only a small number of files were affected.”

If that was true, restore them from a backup or take the hit – how can a small number of files be worth £1M?

Locking down the network after such an attack is a good idea, and this would disrupt office services for certain. But something just doesn’t add up here. It’s possible that the £1M ransom demand has been made up, to cover their embarrassment. Or it could just be sloppy journalism by the BBC – no facts checked and a story about some ransomware being blown out of all proportions. Serious news media haven’t had much to say on the subject. The Register has covered it, but has not repeated the £1M ransom claim.

19-January-1620-January-16

How people get around the Netflix and iPlayer proxy block

Earlier this month at CES, Netflix’s chief product officer Neil Hunt stated that his company’s policy on subscribers accessing content over a VPN remained unchanged. That’s to say that they ask customers not to do it, as it can bust licensing restrictions on content. Neflix is probably the largest provider of streamed TV programmes around the world, now operating in a claimed 190 countries.

I’m not a fan of Netflix – they’re big campaigners for “Net Neutrality”, meaning that all content must be treated the same and ISPs can’t charge more or slow down particular traffic. As their content is not for the public good, and yet accounts for about 40% of the world’s public Internet traffic, they would say that, wouldn’t they? As media organisations such as the BBC (iPlayer) are in the OTT game, the fact that this is a business model where the bulk of the costs are paid for by all Internet users whereas the profits go to the streaming service is not generally mentioned in the popular press. In other words, they profit from the ISP’s investment without contributing anything back. Amazon Prime is another good example.

Anyway, the content that Netflix streams is licensed from content producers, who have good reasons for licensing it on a geographic basis. A TV programme broadcast in one country becomes harder to sell to networks abroad if it’s already available via streaming, and upsetting the status quo won’t be good for content producers. This will leading to less investment in good programming. Netflix is “campaigning” to change this, as though the public, including its customers, have some kind of rights that are being denied. It would, of course, help Netflix’s commercial interests if regional licensing didn’t exist – at least short-term.

That aside, I was amused to see that Neflix’s latest pronouncement, in a company blog post by David Fullagar (VP of Content Delivery Architecture) a week after the CES announcement, that it would now be clamping down on its customers use of proxies or VPNs to smuggle streamed data across boarders. One might surmise that the content providers, many of whom are also local broadcasters, didn’t appreciate Neil Hunt’s complacent sounding comments. The status quo he was defending was basically an weakly enforced contractual prohibition on its customers streaming through a proxy. A actual enforced ban would result in a loss of revenue to Netflix, or if you’re less cynical, would go against the company’s stated aim of “all content free to all (subscribers)”.

But in spite of the soothing words to calm the outrage of its content suppliers, what can Netflix actually do about this? How do you block your customers using a VPN?

It seems to me that it’s impossible to tell whether you’re sending UDP packets to an IP address that’s actually a VPN. It can’t be done. There can be any number of endpoints behind one IP address (an asymmetric NAT LAN), and any number of VPN connections to who-knows-where. And they’ll all appear as one IP address, and the traffic will be indistinguishable.

So how do streaming companies block VPNs now? By having a list IP addresses used by published ones, and that generally means commercial ones. Okay, that might work for the public/commercial VPNs. I shan’t be shedding too many tears if they’re blocked, because they’re making money out of license-busting, which is wrong.

But consider this. Supposing you pay the BBC for a TV license but live abroad for part of the year. You have a moral right to view the content you’ve paid for, and could do so using iPlayer. The only problem is that iPlayer may detect you’re outside the UK by your IP address, and stop you. The solution? Put a proxy server on the network in your house in the UK and connect to it when you’re abroad. I have evidence that this happens a lot.

This can also be done immorally. People in one country with relatives living abroad can set up such a proxy for their friends and relatives to use, and Netflix will be none the wiser. Even if Netflix did suspect an IP address of having too much traffic, what could they possibly do about it? Contact the owner and investigate? How would they even find the owner?

Many ISPs use dynamic addresses in order to charge more for a static one to business customers, with the effect that you don’t know who’s using what IP address today. If you do find a suspected VPN, tomorrow it’s IP address will have changed to one of millions, all used by normal domestic customers.

Finding the many small, private VPNs is going to be impossible. One method might be to probe an IP address to see if a VPN port was open. This is no proof that it’s in use, and no proof that it’s not used for one of the many purposes that a VPN was designed for. And even if they were to try it, it’s simple to restrict access to the VPN ports to your friends abroad. And besides, probing an IP address for an open port without permission is illegal.

The only other method I can think of that would work is to examine the traffic to/from an IP address and see if there’s a correlation between outgoing packets and incoming data from one of Netflix’s servers. But Netflix can’t do that; only an ISP has the technical ability to examine traffic on a particular subscriber’s line. And those are the ISPs that Netflix is abusing by loading them with 40% of their traffic without contributing to the cost. Good luck with that.

14-January-1629-March-16

iZettle contactless payments on American Express (Amex)

Since I reviewed iZettle’s new contactless card reader there have been a few updates to the App, and after the initial teething problems I’m happy to report that it’s been working flawlessly hereabouts.

The latest update is to support contactless payments on American Express. This came as a bit of a surprise, as I assumed it already did! It just goes to show how important Amex is…

You need to do a firmware update. You get this by connecting to your tablet/phone and running the iZettle App. Then go to Settings/Card Readers and select Update. I’ll let someone else try it first, as I can live without the functionality for a while longer.

This does not, of course, work on the freebie iZettle reader – only the Bluetooth one that you pay money for. Don’t be cheap – it’s good!

This update means support for contactless covers Visa, MasterCard, Applepay and Amex. I have to say that I’ve yet to find a card in the UK it couldn’t use, one way or another.

14-January-16

ParentPay won’t support “insecure” browsers

This week that ParentPay, the Microsoftie payment system used by many schools, rolled out a web site update to support an even more limited range of browsers. This included dropping support Internet Explorer before 9 for “security reasons”.

By coincidence, in the same week Microsoft trumped their loyal fanobois at ParentPay by announcing that everything prior to version 10 was hereby deemed unsafe. ParentPay has yet to comment.

However, the notion that any version of Internet Explorer is “safe” is stretching the truth badly. All the mainstream browsers are dodgy; they all support unsafe scripting and embedded code. Microsoft may have the worst reputation, but they’re all undermined by their code and add-ons – and host operating system, to be fair. Only a few niche browsers, that don’t support things like JavaScript and ActiveX, can be considered safe; and those are the ones that ParentPay refuses to support because they don’t allow “rich content”. (And their developers are Microsoft fans). It’s definitely a case of form over security, yet again.

As an illustration of just how feeble their new browser support policy is, here’s a list of those they’ll accept, taken from their web site:

Chrome 35 or higher
Firefox 30 or higher
Internet Explorer 9 or higher
Safari 6 or higher.

The the the the That’s All Folks!

Schools should be seriously considering their relationship with ParentPay, given the cost and inconvenience they’re forcing parents to go through in order to use it. Analysis of the traffic across my servers suggest that IE has around a third of the browser market. Of these, more than half are using IE 9 or earlier.

ParentPay’s assertion that this will only affect a “..small proportion of parents” may be literally true, but it’s disingenuous. Let’s do some simple arithmetic. Say there are 1500 parents in a secondary school. A third of these use IE – that’s 500. Half of these use an old IE (on an old PC) – that’s 250/1500 parents at each school who’ll be grossly inconvenienced. Cancel the fraction out and it’s 1/6, which could be described as a small proportion, but it’s still 250 per school.

The number of people who would be using”unsupported” browsers on tablets or mobile devices is probably very high. Anecdotally, parents have access to a PC somewhere that they currently have to go to in order to use ParentPay. Many would rather use a tablet.

It’s about time someone set up an alternative to ParentPay and schools were educated in to the benefits of open standards.

7-January-167-January-16

Microsoft Security Essentials hangs during a full scan

First off, can I be clear about one thing – endpoint virus scanners don’t make your computer “secure”. A lot of the most dangerous stuff gets past them, but trusting lusers believe they’re safe and will therefore take risks they outerwise wouldn’t. See my posts and academic papers passim ad nauseam. Now that’s out of the way, I favour Microsoft Security Essentials (or Microsoft Endpoint Security) on Windows as I find it less likely to make the system unusable. I don’t recommend it, except as the least-worst option.

On with the problem…

Sometimes, especially in the last year or so, I’ve found Security Essentials will stall when its doing a background scan. You may not notice its done this, but some symptoms are that web browser file downloads won’t work (it’ll download 100% but never finish) and the PC won’t hibernate automatically using the power-saving settings.

I’ve looked for solutions to this, as well as searching the web for an answer. You’ll often see people posting (without references) that this is bug and Microsoft are working on, or have now fixed it. I’ve tried theories myself to see if it’s caused by compression or archive formats causing a decompresser to break (I’ve noticed this often fits the facts), but this is little help when finding a solution, and even then it sometimes still hangs when the option to check compressed files is turned off.

What I’ve yet to find is anyone giving a real solution, so here it is:

Deinstall Security Essentials.
Download and install Security Essentials.

I’ve never known this not to work. On the other hand, I’ve known all the other theories you see posted on forums fail to work pretty consistently.

7-December-157-December-15

How to stop Samba users deleting their home directory and email

UNIX permissions can send you around the twist sometimes. You can set them up to do anything, not. Here’s a good case in point…

Imagine you have Samba set up to provide users with a home directory. This is a useful feature; if you log in to the server with the name “fred” you (and only you) will see a network share called “fred”, which contains the files in your UNIX/Linux home directory. This is great for knowledgeable computer types, but is it such a great idea for normal lusers? If you’re running IMAP email it’s going to expose your mail directory, .forward and a load of other files that Windoze users might delete on a whim, and really screw things up.

Is there a Samba option to share home directories but to leave certain subdirectories alone? No. Can you just change the ownership and permissions of the critical files to root and deny write access? No! (Because mail systems require such files to be owned by their user for security reasons). Can you use permission bits or even an ACL? Possibly, but you’ll go insane trying.

A bit of lateral thinking is called for here. Let’s start with the standard section in smb.conf for creating automatic shares for home directories:

[homes]
    comment = Home Directories
    browseable = no
    writable = yes

The “homes” section is special – the name “homes” is reserved to make it so. Basically it auto-creates a share with a name matching the user when someone logs in, so that they can get to their home directory.

First off, you could make it non-writable (i.e. set writable = no). Not much use to use luser, but it does the job of stopping them deleting anything. If read-only access is good enough, it’s an option.

The next idea, if you want it to be useful, is to use the directive “hide dot files” in the definition. This basically returns files beginning in a ‘.’ as “hidden” to Windoze users, hiding the UNIX user configuration files and other stuff you don’t want deleted. Unfortunately the “mail” directory, containing all your loverly IMAP folders is still available for wonton destruction, but you can hide this too by renaming it .mail. All you then need to do is tell your mail server to use the new name. For example, in dovecot.conf, uncomment and edit the line thus:

mail_location = mbox:~/.mail/:INBOX=/var/mail/%u

(Note the ‘.’ added at the front of ~/mail/)

You then have to rename each of the user’s “mail” folders to “.mail”, restart dovecot and the job is done.

Except when you have lusers who have turned on the “Show Hidden Files” option in Windoze, of course. A surprising number seem to think this is a good idea. You could decide that hidden files allows advanced users control of their mail and configuration, and anyone messing with a hidden file can presumably be trusted to know what you’re doing. You could even mess with Windoze policies to stop them doing this (ha!). Or you may take the view that all lusers and dangerous and if there is a way to mess things up, they’ll find it and do it. In this case, here’s Plan B.

The trick is to know that the default path to shares in [homes] is ‘~’, but you can actually override this! For example:

[homes]
    path = /usr/data/flubnutz
    ...

This maps users’ home directories in a single directory called ‘flubnutz’. This is not that useful, and I haven’t even bothered to try it myself. When it becomes interesting is when you can add a macro to the path name. %S is a good one to use because it’s the name as the user who has logged in (the service name). %u, likewise. You can then do stuff like:

[homes]
     path = /usr/samba-files/%S
     ....

This stores the user’s home directory files in a completely different location, in a directory matching their name. If you prefer to keep the user’s account files together (like a sensible UNIX admin) you can use:

[homes]
     comment = Home Directories
     path = /usr/home/%S/samba-files
     browseable = no
     writable = yes<

As you can imagine, this stores their Windows home directory files in a sub-directory to their home directory; one which they can’t escape from. You have to create “~/samba-files” and give them ownership of it for this to work. If you don’t want to use the explicit path, %h/samba-files should do instead.

I’ve written a few scripts to create directories and set permissions, which I might add to this if anyone expresses an interest.

18-November-1517-March-17

Governments’ hacking fantasies

It’s silly season again.

Yesterday George Osborne warned that Islamists were tooling up and planning deadly cyber-attacks against the UK, targeting critical systems like ATC and hospitals, as he announced government spending on countermeasures would double from about £200M to £400M a year. Mr Osborne shown a rather tenuous grasp of technology in the past, and I fear he’s been watching too many Hollywood movies when forming his current opinion.

I know a bit about ATC, and the chances of a jihadi disrupting NAS over the internet are slight. Damaging aviation is much easier by more direct means.

Likewise, while I have little time for the design of NHS computers systems, even they’d be hard to seriously disrupt. So difficult that it really wouldn’t be worth the bother. If you want to knock out a hospital, blow up the generators and electricity feed – it’s obvious. About the only systemic damage you could do remotely would be to mess up central databases, but these seem to get messed up regularly anyway, and the world goes on.

But this seems positively sane and sensible compared to today’s report from the “US-China Economic and Security Review Commission”. They’re all exercised about those nasty Chinese guys pinching trade secrets by hacking in to US companies and their government agencies. I’m sceptical about the idea that the Chinese government is behind this, and the Commission has weakened the credibility of their claims with their suggested response to the activity:

Yes folks, their suggestion is that Americans hack in to the Chinese systems and steal back or delete the stolen data. How exactly does one steal back data? And do they really think it’s possible to locate, identify and delete stolen data found in a foreign country. Deleting all copies of data from a local system is hard enough, and if the IT department knows its stuff, it’s impossible as it won’t all be on-line.

Whilst there’s plenty of evidence that people in China, and possibly the military, are engaged in cyber-espionage, this idea reads like the plot of another Hollywood movie of the type George Osborne seems to have been watching. Everyone in the security world knows that the majority of criminal activity on the Internet actually comes from…. the USA. This doesn’t mean the US government is behind it – by the sound of the advice they’re getting, they wouldn’t know how.

People like me have been saying that cyber-crime is (going to be) a big problem for many years now, and I welcome governments waking up and taking it seriously at last. The private sector has done spectacularly badly, as the money is in the superficial stuff, and real security gets in the way of profits. It’s just a shame that governments have woken up and are groping groggily around in the dark.

24-October-1517-November-16

TalkTalk Ransom Demand

So, the head of TalkTalk (Dido Harding) has received a ransom demand following the latest hack? From a bunch of Islamist gangsters? I don’t think so. Okay, she probably received an email extortion attempt. Several in fact. It’d be form for Islamist gangsters to have a go, amongst the usual suspects. But the idea that whoever is behind the attacks also sent the ransom demand does not sound like the normal MO. It smells wrong to me. Extortion attempts of this kind generally follow a demonstration that the criminals can disrupt a web site, not after a long-term outage.

I get the vibes that TalkTalk doesn’t know what happened, and take everything they say with a pinch of salt. The only certainty is that their web site was toppled. Data theft, or script kiddies? I suspect the latter, actually. They floated the possibility of widespread data theft, which is very responsible of them until it’s figured out what exactly happened. This is a possibility in any attack.

Meanwhile, people are now questioning whether the stolen data (if there was any) was encrypted, and if not, why not. On a live system, data can’t be encrypted. Think about it! This is allegedly a hack of a live system, so the criminals would have access to the same data that he live system would.

This whole story has been hyped up way beyond the facts. No one (including TalkTalk) wants to suggest it may be overblown for fear of being branded irresponsible by a technically illiterate news media and opportunistic politicians. But it smells all wrong to me. How much more embarrassing if it was was actually script kiddies getting lucky, rather than the APT being hinted at.

23-October-1523-October-15

No talk from TalkTalk

Charles Dunston’s budget ISP TalkTalk has been hacked again. Yawn. This time it’s big news on TV; the headline story in fact. Their website has been KOed for a couple of days, but it’s back online with a front page showing a different news agenda. They get their feed from AOL (also part of the Carphone Warehouse family), who probably just missed the kerfuffle; there’s no celebrity connection after all. Not yet, anyway.

If you’re a TalkTalk retail customer (or possibly a business customer – who knows how their systems interrelate and what data’s been pilfered), and you’ve used the same password with TalkTalk as any other sites, change your password on those sites NOW. The popular media is full of speculation as to what’s been compromised but they’re not mentioning passwords, presumably because TalkTalk will have told them that any passwords would have been encrypted. But if the criminals have got hold of the hashes, which is likely, it’s only a matter of time before they crack them.

How worried should customers of other ISPs be? Pretty worried, as on the serious side of the business they’re known as Opal Telecom, a significant LLU operator providing the link between the last time and the data centre for a large number of Broadband providers.

I can, of course, only speculate as to why this keeps happening to them. One reason might be related to several conversations I’ve had with people from ISPs TalkTalk has taken over along the way. Apparently they really don’t like hard stuff like UNIX/Linux, and within months of a takeover they force a switch to Microsoft before making all the UNIX people redundant. Any fool can use Microsoft – low levels of technical understanding are required, meaning cheap engineers and lower costs. But do their Microsofties actually know what they’re doing? I dare say that some of them do, and some of them don’t. But the bar for a point-and-click Microsoft house going to be lower.