SMBTrapped in Microsoft | TechSNAP 210

chris

7 years ago

Researches find an 18 year old bug in Windows thats rather nasty, we’ve got the details. A new perspective on the bug bounty arms race & the security impact of Wifi on a plane.

Plus great feedback, a bursting round up & much much more!

Thanks to:

Get Paid to Write for DigitalOcean

Direct Download:

RSS Feeds:

Become a supporter on Patreon:

— Show Notes: —

Cylance finds “SPEAR” a new spin on an 18 year old Windows vulnerability

In 1997 Aaron Spangler discovered a flaw in Windows
By causing a user to navigate to a file://1.2.3.4/ url in Internet Explorer, the user’s windows credentials would be sent to the remote server, to attempt to login to it
“Redirect to SMB is a way for attackers to steal valuable user credentials by hijacking communications with legitimate web servers via man-in-the-middle attacks, then sending them to malicious SMB (server message block) servers that force them to spit out the victim’s username, domain and hashed password”
“It’s a serious issue because stolen credentials can be used to break into private accounts, steal data, take control of PCs and establish a beachhead for moving deeper into a targeted network.”
“Software from at least 31 companies including Adobe, Apple, Box, Microsoft, Oracle and Symantec can be exploited using this vulnerability”
“Redirect to SMB is most likely to be used in targeted attacks by advanced actors because attackers must have control over some component of a victim’s network traffic.”
“Less sophisticated attackers could launch Redirect to SMB attacks on shared WiFi access points at locations such as coffee shops from any computer, including mobile devices. We successfully tested this attack on a home network using a Nexus 7 loaded with all required tools.”
“While the user credentials sent over SMB are commonly encrypted, the encryption method used was devised in 1998 and is weak by today’s standards. A stronger hashing algorithm being used on these credentials would decrease the impact of this issue, but not as much as disabling automatic authentication with untrusted SMB servers. With roughly $3,000 worth of GPUs, an attacker could crack any 8-character password consisting of letters (upper and lower case) as well as numbers in less than half a day.”
“Microsoft has yet to release a patch to fix the Redirect to SMB vulnerability. The simplest workaround is to block outbound traffic from TCP 139 and TCP 445 — either at the endpoint firewall or at the network gateway’s firewall (assuming you are on a trusted network). The former will block all SMB communication, which may disable other features that depend on SMB. If the block is done at the network gateway’s firewall, SMB features will still work inside the network, but prevent authentication attempts with destinations outside the network. See the white paper for other mitigation steps.”
“Microsoft did not resolve the issue reported by Aaron Spangler in 1997. We hope that our research will compel Microsoft to reconsider the vulnerabilities and disable authentication with untrusted SMB servers. That would block the attacks identified by Spangler as well as the new Redirect to SMB attack.”
Cylance Whitepaper (PDF)

Given enough money, all bugs are shallow

Eric Raymond, in The Cathedral and the Bazaar, famously wrote: “Given enough eyeballs, all bugs are shallow.”
“The idea is that open source software, by virtue of allowing anyone and everyone to view the source code, is inherently less buggy than closed source software. He dubbed this “Linus’s Law”.”
“However, the Heartbleed SSL vulnerability was a turning point for Linus’s Law, a catastrophic exploit based on a severe bug in open source software. How catastrophic? It affected about 18% of all the HTTPS websites in the world, and allowed attackers to view all traffic to these websites, unencrypted… for two years.”
“OpenSSL, the library with this bug, is one of the most critical bits of Internet infrastructure the world has – relied on by major companies to encrypt the private information of their customers as it travels across the Internet. OpenSSL was used on millions of servers and devices to protect the kind of important stuff you want encrypted, and hidden away from prying eyes, like passwords, bank accounts, and credit card information.”
“This should be some of the most well-reviewed code in the world. What happened to our eyeballs, man?”
“In reality, it’s generally very, very difficult to fix real bugs in anything but the most trivial Open Source software. I know that I have rarely done it, and I am an experienced developer. Most of the time, what really happens is that you tell the actual programmer about the problem and wait and see if he/she fixes it”
“Even if a brave hacker communities to read the code, they’re not terribly likely to spot one of the hard-to-spot problems. Why? Few open source hackers are security experts”
“There’s a big difference between usage eyeballs and development eyeballs.”
“Most eyeballs are looking at the outside of the code, not the inside. And while you can discover bugs, even important security bugs, through usage, the hairiest security bugs require inside knowledge of how the code works.”
Peer reviewing code is a lot harder than writing code.
“The amount of code being churned out today – even if you assume only a small fraction of it is “important” enough to require serious review – far outstrips the number of eyeballs available to look at the code”
“There are not enough qualified eyeballs to look at the code. Sure, the overall number of programmers is slowly growing, but what percent of those programmers are skilled enough, and have the right security background, to be able to audit someone else’s code effectively? A tiny fraction”
“But what’s the long term answer to the general problem of not enough eyeballs on open source code? It’s something that will sound very familiar to you, though I suspect Eric Raymond won’t be too happy about it.”
“Money. Lots and lots of money.”
“Increasingly, companies are turning to commercial bug bounty programs. Either ones they create themselves, or run through third party services like Bugcrowd, Synack, HackerOne, and Crowdcurity. This means you pay per bug, with a larger payout the bigger and badder the bug is.”
However, adding more money to the equation might actually make things worse
“There’s now a price associated with exploits, and the deeper the exploit and the lesser known it is, the more incentive there is to not tell anyone about it until you can collect a major payout. So you might wait up to a year to report anything, and meanwhile this security bug is out there in the wild – who knows who else might have discovered it by then?”
“If your focus is the payout, who is paying more? The good guys, or the bad guys? Should you hold out longer for a bigger payday, or build the exploit up into something even larger? I hope for our sake the good guys have the deeper pockets, otherwise we are all screwed.”
I like that Google addressed a few of these concerns by making Pwnium, their Chrome specific variant of Pwn2Own, a) no longer a yearly event but all day, every day and b) increasing the prize money to “infinite”. I don’t know if that’s enough, but it’s certainly going in the right direction.
“Money turns security into a “me” goal instead of an “us” goal“
“Am I now obligated, on top of providing a completely free open source project to the world, to pay people for contributing information about security bugs that make this open source project better? Believe me, I was very appreciative of the security bug reporting, and I sent them whatever I could, stickers, t-shirts, effusive thank you emails, callouts in the code and checkins. But open source isn’t supposed to be about the money… is it?”
“Easy money attracts all skill levels — The submitter doesn’t understand what is and isn’t an exploit, but knows there is value in anything resembling an exploit, so submits everything they can find.”
“But I have some advice for bug bounty programs, too”:
“You should have someone vetting these bug reports, and making sure they are credible, have clear reproduction steps, and are repeatable, before we ever see them.”
“You should build additional incentives in your community for some kind of collaborative work towards bigger, better exploits. These researchers need to be working together in public, not in secret against each other”.
“You should have a reputation system that builds up so that only the better, proven contributors are making it through and submitting reports”.
“Encourage larger orgs to fund bug bounties for common open source projects, not just their own closed source apps and websites. At Stack Exchange, we donated to open source projects we used every year. Donating a bug bounty could be a big bump in eyeballs on that code.”

FAA Needs a More Comprehensive Approach to Address Cybersecurity As Agency Transitions to NextGen

The Federal Aviation Administration (FAA) faces cybersecurity challenges in at least three areas:
(1) protecting air-traffic control (ATC) information systems,
(2) protecting aircraft avionics used to operate and guide aircraft
(3) clarifying cybersecurity roles and responsibilities among multiple FAA offices
“FAA has taken steps to protect its ATC systems from cyber-based threats; however, significant security-control weaknesses remain that threaten the agency’s ability to ensure the safe and uninterrupted operation of the national airspace systems”
“Modern aircraft are increasingly connected to the Internet. This interconnectedness can potentially provide unauthorized remote access to aircraft avionics systems. As part of the aircraft certification process, FAA’s Office of Safety (AVS) currently certifies new interconnected systems through rules for specific aircraft and has started reviewing rules for certifying the cybersecurity of all new aircraft systems.”
“FAA officials and experts we interviewed said that modern aircraft are also increasingly connected to the Internet, which also uses IP-networking technology and can potentially provide an attacker with remote access to aircraft information systems. According to cybersecurity experts we interviewed, Internet connectivity in the cabin should be considered a direct link between the aircraft and the outside world, which includes potential malicious actors. FAA officials and cybersecurity and aviation experts we spoke to said that increasingly passengers in the cabin can access the Internet via onboard wireless broadband systems.”
“Four cybersecurity experts with whom we spoke discussed firewall vulnerabilities, and all four said that because firewalls are software components, they could be hacked like any other software and circumvented. The experts said that if the cabin systems connect to the cockpit avionics systems (e.g., share the same physical wiring harness or router) and use the same networking platform, in this case IP, a user could subvert the firewall and access the cockpit avionics system from the cabin. The presence of personal smartphones and tablets in the cockpit increases the risk of a system’s being compromised by trusted insiders, both malicious and non-malicious, if these devices have the capability to transmit information to aircraft avionics systems”
One would hope that the cockpit avionics are separated from the onboard entertainment and wifi systems by more than just a firewall. Even if they are not, a properly configured firewall is very difficult to compromise.
Additional Coverage – BatBlue
It seems that the authors of this report were not experts on the subject, and when interviewing experts on the topic, they asked questions like “is there any way to get around a firewall”