Pointers for Mobile App Certificate Pinning

We often get questions about Certificate Pinning in mobile applications. Many clients find the issue difficult to explain to other teams.

You can find really great write ups, and an excellent set of source code examples for fixing this issue – as well as explaining it – at this OWASP.org site.

At a super high level though, you basically want your mobile application to validate the SSL certificate of the specific server(s) that you want it to talk to, and REJECT any certificates that do not match the intended server certificate – REGARDLESS of whether or not the underlying OS trusts the alternative certificate.

This will go a long way to hardening the SSL communication streams between the app and the server, and will not permit easy interception or man-in-the-middle attacks via a network provider or hostile proxy server.

Updates to the app source code are needed to mitigate the issue, and you may need to update apps in the app stores, depending on the way your app is delivered.

As always, if you work with MSI on mobile app security reviews or application-specific penetration testing, we would be happy to demonstrate the attacks and suggested mitigations for any identified issue. Just let us know if you would like assistance.

As always, thanks for reading and I hope your team finds this useful.

Patch for MS15-034 RIGHT NOW!

If you have exposed IIS servers or internal ones as well, pay attention to MS15-034.

Accelerate this patch to immediate. Don’t wait for patching windows, SLAs or maintenance periods. Test the patch, sure, but get it applied ASAP.

This is a remotely executable vulnerability without authentication. It affects a wide range of Windows systems. It offers trivial denial of service exploitation and the bad guys are hard at work building click and drool tools for remote code execution. The clock is ticking, so please, accelerate this patch if possible.

For any additional information or assistance, please contact your account executive or drop us a line via info@microsolved.com.

Thanks and stay safe out there! 

Pay Attention to this Samba Vulnerability

We have a feeling that this recent Samba vulnerability should be at the top of your mind. We are seeing a lot of attention to this across a variety of platforms and we wanted to make sure you saw it. It should be patched as soon as possible, especially on highly sensitive data stores and critical systems.

Let us know if you have any questions.

Heads Up, ICS & SCADA Folks, Especially!

Remotely exploitable vulnerabilities have been identified & published in NTP (network time protocol). This is often a CRITICAL protocol/instance for ICS environments and can be widely located in many control networks. 

The fix currently appears to be an upgrade to 4.2.8 or later.

This should be considered a HIGH PRIORITY for critical infrastructure networks. Exploits are expected as this is an unauthenticated remotely triggered buffer overflow, which should be easily implemented into existing exploit kits.

Please let us know if we can assist you in any way. Stay safe out there! 

Update: 12/19/14 2pm Eastern – According to this article, exploits are now publicly available.

Shellshock: Got Inventory?

Im sure youve all heard of Shellshock by now? If not, its a security flaw in Bash that allows attackers to take control of systems. Bash is really an acronym/pun meaning Bourne-again shellthat was written as a free software replacement for the Bourne shell that preceded it. It is a UNIX shell that acts as a command processor and also reads commands from scripts. The problem is that Bash is present in all kinds of things including Web servers and operating systems. This is a very serious flaw! Worse than any other code vulnerability I can name off hand. There are several serious exploits already extant in the wild. Hundreds of millions of devices and credit cards are at immediate risk of compromise across the globe. Institutions are strongly recommending that people not use their credit cards to make Internet purchases for at least the next several days. Imagine the loss in revenue and buyer confidence this is going to cause! Productivity may well go down and prices may well go up as a consequence of this flaw.

Luckily there are good patches already available to combat this glitch, and I’m sure additional fixes and tweaks are in the offing. But to have any level of safety you need to patch everything on your network that is vulnerable, and you need to do it quickly. Do you know exactly what devices are a part of your network and exactly what operating systems, software and firmware versions are installed on them? Specifically, do you know where Bash is running? If you dont, you may install patches furiously over the next few days and still end up being vulnerable without knowing it. Can you in all good conscience assure your Web customers that their transactions and private information are safe?

Shellshock may have one hidden benefit though; it may be the cold dose of reality that causes organizations to finally get serious about information security and adopt best practices security recommendations, especially where inventories of devices and software are concerned. There is a reason why guidance such as the MSI 80/20 Rule of Information Security and the Top 20 Critical Controls for Effective Cyber-Security list making inventories their number one information security project. If you dont know what you have, how can you possibly secure it?!

Right now, if you are among the prescient few who do keep complete dynamic inventories, ensure that input to all available software fields is validated and have configured each device on your network with a unique admin password, you are sitting pretty! You have the knowledge and time necessary to deal with this problem, and will probably earn kudos and market share from you customers. Isnt that kind of assurance worth spending some time and money on America? 

This blog post contributed by John Davis.

Patch for ShellShock ASAP!

If you haven’t paid attention to the Bash Shellshock vulnerability – NOW IS THE TIME!

Source IPs for probes looking for the vulnerability are growing slowly in number and scope of scans. (As of 9/30/14, 10am Eastern).

There are many vulnerable devices and systems available to exploit and a variety of exploitation vectors exist – including web CGIs, DHCP clients, OpenVPN, SSH, etc. It is highly likely that a wide variety of embedded systems are also vulnerable that meet these capabilities. So far, we have seen attack traffic in the HITME coming from a few SOHO routers and a couple of other embedded network devices. Items like printers, some routers & managed switches, home gadgets, cameras, etc. are likely targets as well.

In the industrial control world, there are a variety of embedded devices leveraging Linux at the core, and many with exposed CGI mechanisms for remote management and monitoring. These need to be inspected as well, as they may also prove vulnerable and potentially exploitable via one or more vectors. Patching may require firmware upgrades in some cases. Contact the vendor for more information.

But, no matter what systems you use and manage, NOW IS THE TIME. Pay attention to this issue and get moving on patching, adding compensating controls and rolling forward with enhanced detection mechanisms. GET BUSY!

As always, if we can assist, feel free to give us a call or drop us a line. We have HoneyPoint emulations for HPSS clients that can help identify sources of traffic and we have assessment signatures for up to the moment known attack vectors. Let us know if we can help!

Thanks for reading, and stay safe out there! 

UPDATE: Good news on Shellshock for embedded devices: If it runs BusyBox, it’s likely NOT vulnerable.

OpenSSL Problem is HUGE – PAY ATTENTION

If you use OpenSSL anywhere, or use a product that does (and that’s a LOT of products), you need to understand that a critical vulnerability has been released, along with a variety of tools and exploit code to take advantage of the issue.

The attack allows an attacker to remotely tamper with OpenSSL implementations to dump PLAIN TEXT secrets, passwords, encryption keys, certificates, etc. They can then use this information against you.

You can read more about the vulnerability itself here. 

THIS IS A SERIOUS ISSUE. Literally, and without exaggeration, the early estimates on this issue are that 90%+ of major web sites and software packages using OpenSSL as a base are vulnerable. This includes HTTPS implementations, many mail server implementations, chat systems, ICS/SCADA devices, SSL VPNs, many embedded devices, etc. The lifetime of this issue is likely to be long and miserable.

Those things that can be patched and upgraded should be done as quickly as possible. Vendors are working on patching their implementations and products, so a lot of updates and patches will be forthcoming in the next few days to weeks. For many sites, patching has already begun, and you might notice a lot of new certificates for sites around the web.

Our best advice at this point is to patch your stuff as quickly as possible. It is also advisable to change any passwords, certificates or credentials that may have been impacted – including on personal sites like banking, forums, Twitter, Facebook, etc. If you aren’t using unique passwords for every site along with a password vault, now is the time to step up. Additionally, this is a good time to implement or enable multi-factor authentication for all accounts where it is possible. These steps will help minimize future attacks and compromises, including fall out from this vulnerability.

Please, socialize this message. All Internet users need to be aware of the problem and the mitigations needed, even for personal safety online.

As always, thanks for reading, and if you have any questions about the issues, please let us know. We are here to help!

Tool Review: Lynis

Recently, I took a look at Lynis, an open source system and security auditing tool. The tool is a local scanning tool for Linux and is pretty popular.

Here is the description from their site:
Lynis is an auditing tool for Unix/Linux. It performs a security scan and determines the hardening state of the machine. Any detected security issues will be provided in the form of a suggestion or warning. Beside security related information it will also scan for general system information, installed packages and possible configuration errors.

This software aims in assisting automated auditing, hardening, software patch management, vulnerability and malware scanning of Unix/Linux based systems. It can be run without prior installation, so inclusion on read only storage is possible (USB stick, cd/dvd).

Lynis assists auditors in performing Basel II, GLBA, HIPAA, PCI DSS and SOx (Sarbanes-Oxley) compliance audits.

Intended audience:
Security specialists, penetration testers, system auditors, system/network managers.

Examples of audit tests:
– Available authentication methods
– Expired SSL certificates
– Outdated software
– User accounts without password
– Incorrect file permissions
– Configuration errors
– Firewall auditing 

As you can see, it has a wide range of capabilities. It is a pretty handy tool and the reporting is pretty basic, but very useful.

Our testing went well, and overall, we were pleased at the level of detail the tool provides. We wouldn’t use it as our only Linux auditing tool, but is a very handy tool for the toolbox. The runs were of adequate speed and when we tweaked out the configs with common errors, the tool was quick to flag them. 

Overall, we would give it a “not too shabby”. 🙂 The advice is still a bit technical for basic users, but then, do you want basic users administering a production box anyway? For true admins, the tool is perfectly adequate at telling them what to do and how to go about doing it, when it comes to hardening their systems.

Give Lynis a try and let me know what you think. You can give me feedback, kudos or insults on Twitter (@lbhuston). As always, thanks for reading! 

Java 0-Days are Changing Corporate Use Patterns

With all of the attention to the last few Java 0-days and the market value for them falling them (which many folks believe indicate there are more out there and more coming), we are starting to hear some organizations change their policies around Java, in general. 

It seems some clients have removed it from their default workstation images, restricting it to the pile of as-needed installs. A few have reported requiring more frequent Java update settings and a couple have talked about switching in-house development away from Java to different languages. 

Is your organization changing the way you view Java? How are things changing around the IT shops you work with? 

Drop us a line in the comments or via Twitter (@microsolved or @lbhuston) and let us know what YOU think!