Slashdot Mirror

What shocks me in all of these e-mail leak scandals is how un verified it is. I remember being able to telnet to open port 25s and send e-mail to anyone as anyone. PGP encryption and signing should be standard by anyone at that level.

There were DKIM signatures on the Hillary Clinton emails

http://blog.erratasec.com/2016...

This Politifact post muddles over whether the Wikileaks leaked emails have been doctored, specifically the one about Tim Kaine being picked a year ago. The post is wrong -- we can verify this email and most of the rest.

In order to bloc spam, emails nowadays contain a form of digital signatures that verify their authenticity. This is automatic, it happens on most modern email systems, without users being aware of it.

This means we can indeed validate most of the Wikileaks leaked DNC/Clinton/Podesta emails. There are many ways to do this, but the easiest is to install the popular Thunderbird email app along with the DKIM Verifier addon. Then go to the Wikileaks site and download the raw source of the email https://wikileaks.org/podesta-....

As you see in the screenshot below, the DKIM signature verifies as true.

If somebody doctored the email, such as changing the date, then the signature would not verify. I try this in the email below, changing the date from 2015 to 2016. This causes the signature to fail.

Robert does a nice job debunking Tim Wu's editorial. http://blog.erratasec.com/2017...

Update: No, it's probably not okay to call people "morons" as Linus does. They may be wrong, but they usually are reasonable people. On the other hand, security people tend to be sanctimonious bastards with rigid thinking, so after he as dealt with that minority, I can see why Linus treats all security people that way.

Like Robert Graham describes in http://blog.erratasec.com/2017..., it's a type of attack that can be perpetrated against any service on the internet.

Solutions:
- Build a proxy service (per the article) that parses input before passing it to $SERVICE.
- Do not put it on the internet (i.e. firewall).

Is SMB open by default in Windows Firewall anyway? If anything, pooh-pooh Redmond for that. I know, I know, millions of affected hosts.

even though AFAIC USD is inherently flawed and failing as money, it is still more stable in intraday trading so far, it doesn't jump up and down by 20% in a day

Then again, there was that one day when the dollar lost nearly 70% of its value.

But that was an intentional act of government. Shifts in the value of bitcoin happen by market forces.

The "price" of bitcoin is the result of people trying to buy and sell it. Naturally, the bigger the market, the less effect any one transaction has on the price. Dollar markets are gigantic, so the value doesn't move a whole lot from day to day. Bitcoin markets are small, just barely big enough to be considered small by the standards of financial professionals. A few million dollars can push the price around a fair bit. And because the markets are small, people don't have the capital to allocate to price points much beyond the expected range, so a breakout can overrun quite a bit before settling back down.

In short, the swings in the price of bitcoin are the result of you being born when you were, and not because bitcoin has any sort of inherent instability. If you had been born a few decades later, either you'd see a much larger and more stable bitcoin market, or you'd only ever see the name in the history books.

Ethereum seems to be different, by the way. I haven't been following it closely, but it isn't just another Dogecoin or other clone-of-bitcoin-with-different-magic like used to pop up so often. It is trying to do something completely different, and in my opinion, it is far too early to tell if it will succeed in doing so, or for that matter, if that different thing is a good idea or not. There have been problems encountered along the way, to be sure.

How much of the interest in Ethereum is because it is special, and how much is because it is new(er) - I have no idea.

It's good if the information is accurate.

Accuracy is always good, but a lot of it is subjective. And, besides, when Hillary's e-mails were posted, no one protested the content. People were outraged over "Putin" meddling in the US elections, but I don't recall anyone calling any particular e-mail a fake...

Hillary's E-mails were verified by their DKIM headers. As far as I'm aware, none of those E-mails have been shown as fake, and it would be pretty hard to fake the DKIM signature or claim that an E-mail was fake if it had a correct DKIM signature.

I don't know if a similar mechanism is available here, we'll probably find out in the next day or so.

Overall, I'm completely in favor of any correct (verified, impossible to fake) data dump on political candidates, including Trump and other GOP leaders if there were any.

We've always said "if you don't want people to know what you're doing, don't put it on the internet". What's good for the people is good for the leaders. We need to show them what it's like to have shit for privacy, like us.

This sort of thing will only make cheating and corruption harder for future candidates. Knowing that any aide or sysadmin could make their innermost decisions public lowers the "iiquidity" of such actions quite a bit. Even cell phone conversations cannot be considered private any more.

Data dumps are currently spotty and one-sided, but I expect future big elections will show both sides in stark.

And that's a good thing.

First, you apparently don't know what I believe. Second, it's really odd that you believe nothing will change when most people against Trump are saying silly things about Nazis, martial law and nuclear annihilation. Third, rationality doesn't work that way. You believe people based on evidence.

That said, can you explain exactly why it's unreasonable to have a low Bayesian prior for P(they're lying) when they've repeatedly lied, even to cause wars, in the past, when they've presented ridiculous and flimsy evidence that was torn apart, and the best they can give us are political hacks and anonymous sources? Or perhaps you can explain why empiricists should believe things without evidence after having been given quite a lot of disproved evidence?

Oh wait, no, you were just building a straw man ... Is Burning Man coming up again? For some reason it feels that way...

Robert Graham explained it succinctly: http://blog.erratasec.com/2017... .

The real story here is that Giuliani is now a goddamn cybersecurity advisor, not that this personal site is crap. The guy was hired not because of competence but because he spent the entire campaign kissing Trump's ass.

If that's how Trump works, then why didn't he give Giuliani what he wanted - the Secretary of State job? It would have saved Trump a lot of grief that he's going thru w/ Rex Tillerson, and Rudy would have got his first choice

This role would probably not have been created so quickly, had the Dems not been obsessed w/ the Russians - something they conveniently ignored the last 8 years

Robert Graham explained it succinctly: http://blog.erratasec.com/2017... .

The real story here is that Giuliani is now a goddamn cybersecurity advisor, not that this personal site is crap. The guy was hired not because of competence but because he spent the entire campaign kissing Trump's ass.

"Thus historian Vincent J. Cannato concluded in September 2006, "With time, Giuliani's legacy will be based on more than just 9/11. He left a city immeasurably better off — safer, more prosperous, more confident — than the one he had inherited eight years earlier, even with the smoldering ruins of the World Trade Center at its heart. Debates about his accomplishments will continue, but the significance of his mayoralty is hard to deny."

You might be correct, in that Giuliani was not hired because of competence, but you are completely incorrect implying that Giuliani is wholly without competance.

And once again, I have to ask: is [what you said] this important? Is *why* someone is hired more important than their competence?

And once again again, I have to ask: compared to what? Is hiring Giuliani any worse than the practices of the previous administration or the runner-up candidate?

For contrast, note that Bush appointed a crony as head of FEMA who completely fell on his face during Katrina, and Obama appointed Caroline Kennedy as ambassador to Japan, who was completely outmastered in our recent Japanese treaty negotiations(*).

Is it useful *at all* to just throw throws random aspersions around?

(*) Resulting in a treaty which is beneficial to Japan, but a very bad deal for America. I have no opinion about Ms. Kennedy, good or bad, only note that she was unqualified for the position, was apparently appointed because of her ties to a famous family dynasty, and America was worse off because of it.

> Giuliani has been hired to endorse and push laws that further Trump's administration's ability to invade the privacy of those they dislike, and to prosecute those who dare to use technology or the internet to speak out against them.

Doesn't have to, Obama already enacted that for him, including the information sharing directive from the other day.

Looks like they picked a dumb hosting provider, but there was nothing worthwhile there to hack. It certainly does look bad if they picked a host that bad, though.

Robert Graham explained it succinctly: http://blog.erratasec.com/2017... .

The real story here is that Giuliani is now a goddamn cybersecurity advisor, not that this personal site is crap. The guy was hired not because of competence but because he spent the entire campaign kissing Trump's ass.

Well that, plus his business ties with the Russians.

Robert Graham explained it succinctly: http://blog.erratasec.com/2017... .

The real story here is that Giuliani is now a goddamn cybersecurity advisor, not that this personal site is crap. The guy was hired not because of competence but because he spent the entire campaign kissing Trump's ass.

> That only confirms the headers. The bodies of the emails are Russian fabrications.

Okay, so click here and then the "view source" link and you can read the DKIM signature yourself. I'll save you some trouble and copy paste it:

DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;
                d=hillaryclinton.com; s=google;
                h=from:mime-version:references:in-reply-to:date:message-id:subject:to :cc;
                bh=EHIyNFKU1g6KhzxpAJQtxaW82g5+cTT3qlzIbUpGoRY=;
                b=JgW85tkuhlDcythkyCrUMjPIAjHbUVPtgyqu+KpUR/kqQjE8+W23zacIh0DtVTqUGD
                  mzaviTrNmI8Ds2aUlzEFjxhJHtgKT4zbRiqDZS7fgba8ifMKCyDgApGNfenmQz+81+hN
                  2OHb/pLmmop+lIeM8ELXHhhr0m/Sd4c/3BOy8=
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;
                d=1e100.net; s=20130820;
                h=x-gm-message-state:from:mime-version:references:in-reply-to:date :message-id:subject:to:cc;
                bh=EHIyNFKU1g6KhzxpAJQtxaW82g5+cTT3qlzIbUpGoRY=;
                b=dEYKdN2vH085sl/02zUgJ1Lr66LV8lRV9Lrqx9SIpfiF1bOLLbIr1Au6AAY5vwg1vS
                  klK/TvacKT0j8aYADGNWP6BtG5XZ+IME6ydojlufQ3jqksqLkycSJ2ahYhxw4LmCii8n
                  kja2EKzRFcKGPnfhYnfwBCmIk/D5FWN6+yvpAYSmmZlxsR4b7mTJ8r/NmB7dKRIHeq8b
                  Ersjyl8edCTfC6nGbUrEEV7C6uQE3N16B5m2XPnRATWSuWj/Nz7ZsM/9snj+rlTjJx5e
                  wI5Epet9ADtlAWqJw/L/5HCNaAFqyR3QK1/AFjsTk+Q2METC3+0Eo+yMaArw2viFZLu4
                  hvoQ==

What does that mean? Let's check Wikipedia:

The DKIM-Signature header field consists of a list of tag=value parts. Tags are short, usually only one or two letters. The most relevant ones are b for the actual digital signature of the contents (headers and body) of the mail message, bh for the body hash, d for the signing domain, and s for the selector. The default parameters for the authentication mechanism are to use SHA-256 as the cryptographic hash and RSA as the public key encryption scheme, and encode the encrypted hash using Base64.

Now, would you like to go back and look at the b and bh parameters in the signature and tell me what those mean? Right, they cover contents (headers and body) as well as the body hash. If you want to make a serious claim that this is fake, give me a link to the blockchain transaction when you win 1 BTC from Erratasec for breaking DKIM.

I'm waiting.

Here's a nice little summary video:
https://www.youtube.com/watch?...

Want to claim they're fake? Give me the blockchain transaction when you win this challenge for 1 BTC:
http://blog.erratasec.com/2016...

You do not, because you cannot, argue with this. You just post insults. Because that's all you can do. You will not, because you cannot, argue against any of the things found in the email. You just ignore them.

> If not 100% buying into info obtained illegally with no means of detecting which emails (in entirety) might have been planted maybe.

You can perform DKIM validation on the emails. Erratasec will give you a bitcoin if you can spoof them. The simplest way to settle this is to show me a link to the blockchain transaction when you win.

"This is nonsense. The evidence available on the Internet is that Trump neither (directly) controls the domain "trump-email.com", nor has access to the server. Instead, the domain was setup and controlled by Cendyn, a company that does marketing/promotions for hotels, including many of Trump's hotels. Cendyn outsources the email portions of its campaigns to a company called Listrak, which actually owns/operates the physical server in a data center in Philidelphia."

http://blog.erratasec.com/2016...

It's amazing that you think that someone, since at least 2009, has been sending fake messages with fake senders from Gmail, Yahoo and clintonemail.com without anyone wondering about the oddly incriminating messages showing up in their inboxes.

If you think you can fake these, you can get 1 bitcoin from erratasec.

Btw, if you can forge an email that validates correctly as I've shown, I'll give you 1-bitcoin. It's the easiest way of solving arguments whether this really validates the email -- if somebody tells you this blogpost is invalid, then tell them they can earn about $600 (current value of BTC) proving it. Otherwise, no.

Please show me the blockchain transaction when you win this, ok?

All that really matters is whether the leaks are true. In the case of Podesta's emails, we can validate them by the DKIM signature. We know how Podesta was hacked, though, it was exactly this phishing email. See how that is a bitly link to "reset" his Google password? What I don't know is how they could possibly miss that GMail would flag this saying something like "THIS IS NOT FROM GOOGLE YOU MORON."

Similarly, whether or not that was American intelligence that hacked a top Russian aide, what's perhaps more worrying is that Hillary has been pushing for a no-fly-zone that would leave us shooting down Russian planes over Syria. To most people who know about the area, this sounds like a great way to start WWIII. This might help explain why she's so eager to to expand the draft so women can #FightForHer.

This builds a good case that (a) no NSL or gag order was given, (b) the hardware used is commodity, and therefore the actual build is a matter of tinkering, not invention, and (c) the build is legal* with regard to radio frequency encryption and the computer fraud and abuse act

*Read the article for more nuance.

So you've personally audited it and certified that in your capacity as an AC Tor expert?

Actually, the configuration and logs of the site was made publicly available during the trial. People recreating the configuration the government claimed the server had do not get a public IP address when they connect to the server in the way the government claims they connected to the server, they get a MySQL admin page. DPR screwed up the configuration, but the government lied -- sorry, misspoke -- about how they located the IP of the server.

Agreed. And in fact one guy narrowed down the source of the packet injection to the Great Firewall (China Unicom): Pin-pointing China's attack against GitHub.

It injects advertising into search engine results, and also has the capability to intercept and hijack SSL/TLS connections to websites, thanks to the installation of a self-signing certificate authority on affected machines.

It's worst than that. Not only can the program MITM SSL/TLS connections on the infected machine, so can anyone else in a position to intercept the traffic. The private signing key employed by the program is public knowledge at this point, and the same on all infected systems.

See http://blog.erratasec.com/2015...

Now all these boxes can be owned by anyone with the key!

FUD? Or you plugging your ears and shouting "LA LA LA NOT LISTENING"?

The fact is it's exploitable in the wild and this is just a quick-and-dirty PoC scan that was looking for vulnerable servers by requesting the root page.

Enumerating all the exploitable PHP scripts is intractable - the vector is just plain old CGI standard populating environment with HTTP headers. If the script uses something like system() for some reason and calls something that is, or calls, a bash script, it can trigger the vulnerability.

Sadly, this is not the case. The evidence is that bad actors had this exploit for months: http://arstechnica.com/securit...

One of the two sites cited as evidence have since taken a step back,

Important update (10th April 2014): Original content of this blog entry stated that one of our SeaCat server detected Heartbleed bug attack prior its actual disclosure. EFF correctly pointed out that there are other tools, that can produce the same pattern in the SeaCat server log (see http://blog.erratasec.com/2014... ). I don't have any hard data evidence to support or reject this statement. Since there is a risk that our finding is false positive, I have modified this entry to neutral tone, removing any conclusions. There are real honeypots in the Internet that should provide final evidence when Heartbleed has been broadly exploited for a first time.

This guy has retracted part of his analysis based on comments, but tries to make a case that passwords and cookies in the http headers are more likely to be exposed than keys. Remember, http-auth is still used a lot. http://blog.erratasec.com/2014...

For people who didn't follow the link chain, it has since been updated:

Important update (10th April 2014): Original content of this blog entry stated that one of our SeaCat server detected Heartbleed bug attack prior its actual disclosure. EFF correctly pointed out that there are other tools, that can produce the same pattern in the SeaCat server log (see http://blog.erratasec.com/2014... ). I don't have any hard data evidence to support or reject this statement. Since there is a risk that our finding is false positive, I have modified this entry to neutral tone, removing any conclusions. There are real honeypots in the Internet that should provide final evidence when Heartbleed has been broadly exploited for a first time.

And there's already a comment there: http://blog.erratasec.com/2014...

Have a look at BadBIOS http://blog.erratasec.com/2013/10/badbios-features-explained.html, which infects BIOS by merely inserting a reprogrammed USB flash drive. It works on OS X, which is a certified Unix. And to know how easy it is to reprogram the firmware of a managed flash (SD card, USB flash drives), check this post from few days ago http://www.bunniestudios.com/blog/?p=3554

All released Linux versions tried to be bug free, that should be nothing as big to deserve a whole new version for 4.0. But probably this "bug fix" goes beyond the normal scope. It must not just work, but work in an hostile environment where governments with plenty of resources try to exploit any "more or less work" vulnerability to plant backdoors and snoop, where hardware, firmwares (the methods that could use #badBIOS to spread could be an example), internet protocols or encryption algorythms are not so trustable, and could had been malicious commits in the past, not just work, but work even with a legion of high profile attackers trying to find some potential hole to sneak in.

Robert Graham has published a well-written response:

http://blog.erratasec.com/2013/10/badbios-features-explained.html

What about hardware backdoor activation? There had been rumors of intel putting 3G radios in vPro cpus, and there had been backdoors in FPGAs. There had been a nice presentation in DEFCON17 around this topic.

Do you trust your exit node or proxy? Defcon had a recent talk on setting up proxy servers as a very quick way to find people who have something to hide. Now you have their IP address and their destination.

It's not just about exit nodes anymore. The NSA can, and regularly does, de-anonymize users within the Tor network, with or without compromised or 'baddie-controlled' exit nodes.

Tor works only as long as exit nodes are not in the bad guy's control.

Correction: Tor only works (in its current implementation) when there isn't a single bad node in the entire network. IOW, not going to happen.

Let's also keep in mind that 60+% of the funding for Tor, comes directly from the Department of Defense (DoD).

Concerned yet? You should be.

The important bit, the one that has value to *me*, is that it can hide my identity. It can hide the identity of people who are afraid of oppression, it can hide the identity of whistle blowers, it can hide the identity of people asking for help.

Actually, no, it can't. You're thinking of i2p, not Tor.

If you're confused, you need to read up on the major flaws and vulnerabilities in Tor that allow the NSA or enough controlling entities to de-anonymize anyone using Tor. In fact, the more Tor exit nodes, the easier it is.

The tremendous spike in users using Tor could be both in reaction to the NSA news, or proactive from the NSA ramping up their use of Tor to more-rapidly de-anonymize the traffic coming across those exit nodes.

Read and enjoy, and share with others:

http://blog.erratasec.com/2013/08/anonymity-smackdown-nsa-vs-tor.html

They don't need to crack the traffic, if they de-anonymize the clients in and out of Tor, it's cleartext on both ends. With SSL and VPNs now being as wide open as cleartext and easily broken, it's a moot point.

http://blog.erratasec.com/2013/08/anonymity-smackdown-nsa-vs-tor.html

And direct yourself and others to Project Meshnet:

https://projectmeshnet.org/

Sorry, but it's now too easy to de-anonymize traffic within and around Tor. The more Tor nodes there are, the -easier- it becomes.

Read and learn, then direct everyone to start using i2p and Project Meshnet (https://projectmeshnet.org/)

http://blog.erratasec.com/2013/08/anonymity-smackdown-nsa-vs-tor.html

You estimate of Tor's privacy is higher than mine, and, evidently, PJ's.

An interesting read, thanks! What about setting up Freenet's Freemail then?

You estimate of Tor's privacy is higher than mine, and, evidently, PJ's.

http://www.erratasec.com/ferret.html

This might be believable, except this is the guy who came up with the bogus apple wireless exploit a while back. They reported it to Apple, but it wasn't really a problem because it was with a third-party wireless device, in a setup that would probably never happen in real life (and likely was never actually a vulnerability in the first place). Furthermore, if you are willing to pay for their Hacker Eye View program penetration testing, they WILL give you a full report of the vulnerability. After a careful analysis of the situation, it seems that the 'security expert' is actually a marketoid looking to drum up some free publicity for his company.