I'm enhancing a commercial App which until now has used cloud AI models to analyse data and make predictions.
The enhancement is moving the models onto the app for applications with no or limited network access.
These models represent significant IP to our clients and it is essential that we secure any data downloaded to a device from theft.
The App is iOS only for now and I was intrigued by WWDC2020's CoreML update including support for encrypting models. This would be ideal but we can't use CoreML at the moment due to its API not supporting the methods our models require.
Nice to know though that this is a recognised issue with in-app ML model usage.
What is the best method and available options in iOS (>11.0) right now that won't run foul of encryption export laws or even Apple's app store rules etc?
Or models are Javascript which we run in a JavaScriptCore VM with additional data files loaded from json string files.
My current thinking is to use something like the iOS AES encryption. Not hardwire the private key in the app but instead pass it via https, after a user logs in, storing it in the keychain. Decrypt the data strings in memory before loading into the JS VM.
I can see the obvious weaknesses with this approach and would be keen to hear how others have approached this?
The Data
The enhancement is moving the models onto the app for applications with no or limited network access.
These models represent significant IP to our clients and it is essential that we secure any data downloaded to a device from theft.
From the moment you make the data/secrets public, in the sense you include it with your mobile app binary or later download it into the device and store it encrypted, you need to consider it compromised. No bullet proof around this, no matter what you try, you can only make it harder to steal, but with all the instrumentation frameworks available to introspect and instrument code at runtime, your encrypted data can be extracted from the function that decrypts it:
Decrypt the data strings in memory before loading into the JS VM.
An example of a very popular instrumentation framework is Frida:
Inject your own scripts into black box processes. Hook any function, spy on crypto APIs or trace private application code, no source code needed. Edit, hit save, and instantly see the results. All without compilation steps or program restarts.
The Private Key
My current thinking is to use something like the iOS AES encryption. Not hardwire the private key in the app but instead pass it via https, after a user logs in, storing it in the keychain.
While not hard-coding the private key in the device is a wise decision it doesn't prevent the attacker from performing a man in the middle(MitM) attack to steal it, or use an instrumentation Framework to hook into the code that stores it in the keychain, but you may already be aware of this or not, because it's not clear from:
I can see the obvious weaknesses with this approach...
In my opinion, and as a side note, I think that first you and the business need to consider if the benefits for the user in having the predictions being made locally on their device outweighs the huge risk being taken of moving the data from the cloud into the device, and data protections laws need to be taken in consideration, because the fines when a data breach occurs can have a huge impact in the organization future.
iOS Solutions
What is the best method and available options in iOS (>11.0) right now that won't run foul of encryption export laws or even Apple's app store rules etc?
I am not an expert in iOS, thus I cannot help you much here, other then recommending you to use as many obfuscation techniques and run-time application self-protections(RASP) in top of the solution you already devised to protect your data, so that you can make an attacker life harder.
RASP:
Runtime application self-protection (RASP) is a security technology that uses runtime instrumentation to detect and block computer attacks by taking advantage of information from inside the running software.
RASP technology is said to improve the security of software by monitoring its inputs, and blocking those that could allow attacks, while protecting the runtime environment from unwanted changes and tampering.
You can also try to use advanced bio-metrics solutions to ensure that a real user is present while the mobile app is being used, but bearing in mind that the more skilled attackers will always find a way to extract the data to a command and control server. It's not a question if they will be able, but when it will happen, and when it happens it's a data breach, and you need to have planned ahead to deal with it's business and legal consequences.
So after you apply the most suitable in app defenses you still have an issue left to resolve, that boils down to ensure your API server knows what is making the request, because it seems you already have implemented user authentication to solve in behalf of who the request is being made.
The Difference Between WHO and WHAT is Accessing the API Server
When downloading the data into the device you need to consider how you will ensure that your API server is indeed accepting the download requests from what you expect, a genuine instance of your mobile app, not from a script, bot, etc., and I need to alert you that user authentication only says in behalf of who the request is being made, not what is doing it.
I wrote a series of articles around API and Mobile security, and in the article Why Does Your Mobile App Need An Api Key? you can read in detail the difference between who and what is accessing your API server, but I will extract here the main takes from it:
The what is the thing making the request to the API server. Is it really a genuine instance of your mobile app, or is it a bot, an automated script or an attacker manually poking around your API server with a tool like Postman?
The who is the user of the mobile app that we can authenticate, authorize and identify in several ways, like using OpenID Connect or OAUTH2 flows.
Think about the who as the user your API server will be able to Authenticate and Authorize access to the data, and think about the what as the software making that request in behalf of the user.
I see this misconception arise over and over, even among experienced developers, devops and devsecops, because our industry is more geared towards identifying the who not the what.
Others approach
I can see the obvious weaknesses with this approach and would be keen to hear how others have approached this?
As I said previously I am not an expert in iOS and I don't have more to offer to you then what I have already mention in the iOS Solutions section, but if you want to learn how you can lock your mobile app to the API server in order to only reply with a very high degree of confidence to requests from a genuine instance of your mobile app, then I recommend you to read my accepted answer to the question How to secure an API REST for mobile app?, specifically the section Securing the API server and the section A Possible Better Solution, where you will learn how the Mobile App Attestation concept may be a possible solution for this problem.
Do you want to go the Extra Mile?
In any response to a security question I always like to reference the amazing work from the OWASP foundation.
For Mobile Apps
OWASP Mobile Security Project - Top 10 risks
The OWASP Mobile Security Project is a centralized resource intended to give developers and security teams the resources they need to build and maintain secure mobile applications. Through the project, our goal is to classify mobile security risks and provide developmental controls to reduce their impact or likelihood of exploitation.
OWASP - Mobile Security Testing Guide:
The Mobile Security Testing Guide (MSTG) is a comprehensive manual for mobile app security development, testing and reverse engineering.
For APIS
OWASP API Security Top 10
The OWASP API Security Project seeks to provide value to software developers and security assessors by underscoring the potential risks in insecure APIs, and illustrating how these risks may be mitigated. In order to facilitate this goal, the OWASP API Security Project will create and maintain a Top 10 API Security Risks document, as well as a documentation portal for best practices when creating or assessing APIs.
Related
I am working on a project that will include a Laravel web application, Android application, and iOS application. I am new to iOS. Since a user's data will need to be synchronized across all platforms, I assume the optimal way to do this is with a Relational Database (i.e. MySQL) and a PHP Web Service that interacts with the mobile apps.
Everything I see online says to basically use Firebase for mobile apps. There is a lot of work to create the PHP Web Services necessary. I am surprised the dominant recommendation is to use Firebase. About every iPhone app I have has both web and mobile platforms where data must be synchronized. I assume they cannot use Firebase and must use PHP Web Services.
Am I missing something?
You are absolutely correct that it is not a foregone conclusion that one would use Firebase. It’s a great solution that fills a particular niche, but developing separate web services is extremely common. (I suspect that that your online research may be biased by the fact that most of these articles are likely geared for mobile developers who might not have the wherewithal to develop, maintain, and support their own web server infrastructure.)
That having been said, I would not have jumped to the conclusion that the “optimal” way would be to write your own backend going against some RDBMS. These NoSQL solutions, like Firebase, are perfectly up to the job in most cases. There are pros and cons on both sides of the NoSQL vs RDBMS discussion (which is probably beyond the scope of this question).
So, do not be unduly biased by these Firebase articles you found online, but consider adding it to your tech stack if:
there are some compelling Firebase features that you don’t want to reinvent yourself (e.g. the integrated authentication options are nice; the realtime observers for database changes is a killer feature, if you need that; etc.);
the NoSQL paradigm of Firebase’s “Real-time Database” fits your app’s requirements; and
you need backend server capabilities, but don’t want to deal with the overhead of developing and maintaining your own backend.
In your case, because you’re already developing a Laravel app, that largely undermines that last rationale, because you’ve obviously already signed up for that.
So, it is just a question of what Firebase features you need and whether these features are compelling enough to justify adding Firebase to your tech stack. But don’t use Firebase because you found a bunch of articles advocating for it. Nor should you dismiss Firebase because you fear it won’t be “optimal”. It depends.
All of that having been said, the excellent, seamless, object-to-database mapping that Laravel provides really begs for the SQL database approach. If you tried to use Firebase for the backend, you’d likely lose a lot of the benefits of Laravel.
I have to develop library for 3rd party. This library have to be secure so that only parties that have credentials can use it.
I am going in a direction that I will provide API key that 3rd party app will have to enter it in order for library to work.
Is there any possibility that I do some sort of locking of a Cocoapod? Or is Framework better solution for this kind of problem?
Does anyone have any other solution/suggestion?
It would be better if you elaborated a bit on what exactly you are trying to achieve and what type of task this library performs, maybe give some examples. If I understand you correctly, you need to prevent usage of library by those who did not pay for the service etc.
The key approach will be OK if the library is a client for some webservice. In this case, you should have API keys anyway to protect the API itself, so the client library will just forward this key to the webservice. This approach is widely used in lots of client libraries.
If the library does work only locally (for example, it performs some science-heavy computation / computer vision / etc), then you can just give out the compiled library and license to those who have already paid. You can protect it with a key of course, but it is not too useful, as the key will likely be validated locally, therefore it can easily be compromised or reverse engineered. So the only good way will be to distribute the library to those who purchased / requested trial, and force upon them a license which will restrict the library's usage.
EDIT
If by "Cocoapod" you mean "distributing as source code" and by "Framework" you mean "distributing as a binary", then it depends on what exactly you do in the library. If it is just connecting to the endpoint and marshalling data (e.g. parsing), you can just distribute the source version, as there is no "know-how" to that. On the other hand, if there is something business-related and specific done besides contacting the API, use closed source distribution (binary).
Source distribution has a benefit that you don't have to recompile it if new target architectures appear. It is also easier to distribute via CocoaPods, and your library users will like it more (for a variety of reasons).
I am working on an app. Say, it should be secure and safe for the end user, to the degree of a matter of life and death, in the most extreme case. In reality, it's not so hard but, let's assume it.
Thus, I want to make sure, that if serious bad guys get this iPhone and do their tricky work to disassemble it, jailbreak, whatever to get the data from the app, then they get as least clue as possible.
I want to build, test the app and its environment the safest way.
The questions are:
Are there official tools from Apple or other sources to test not
only the app itself but all the security stuff?
How much should I be worried about bad guys gaining access to the
filesystem? How can I prevent data revealing?
How reliable, e.g. backdoorless are existing encryption libraries?
For help with security testing an iOS app, I would recommend checking OWASP's Mobile Security Project. There are a lot of resources about common vulnerabilities in mobile applications, but also guidance on the steps to test a mobile application.
For your specific questions:
XCode has a built-in Analyze feature that looks for problems within the source code of your application. This is a form of static analysis. There are third-party tools that help with dynamic analysis, testing the running application. OWASP ZAP and Burp Suite are examples of tools in this category.
If a user has a jailbroken phone, they'll like have access to the whole filesystem. It's also not possible to protect completely against reverse engineering. This post from the Information Security community might be helpful in that regard. You can however limit the sensitive information you store on the device. Be careful about what information is stored in log files, cached files, plist files, basically anything stored on the device. If the information is very sensitive, it might be better to store it on the server rather than device, since you own the server and don't have direct control over a user's device.
I would consult the Developer's Guide to Encrypting and Hashing Data as well as the iOS Security Guide. I don't know about specific encryption libraries, but in general the most common problem is poor implementation of encryption libraries rather than problems with the libraries themselves. Also, generally using existing libraries is a better practice than trying to create your own.
I'd also consult the Information Security Community, they'll have more guidance on how to security test iOS applications.
I recently read about decompilation of iOS apps and I'm now really concerned about it. As stated in the following posts (#1 and #2) it is possible to decompile an iOS which is distributed to the App Store. This can be done with jailbreak and I think with copying the app from memory to hdd. With some tools it is possible to
read out strings (strings tools)
dump the header files
reverse engineer to assembly code
It seems NOT to be possible to reverse engineer to Cocoa code.
As security is a feature of the software I create, I want to prevent bad users from reconstructing my security functions (encryption with key or log in to websites). So I came up with the following questions:
Can someone reconstruct my saving and encryption or login methods with assembly? I mean can he understand what exactly is going on (what is saved to which path at which time, which key is used etc., with what credentials is a login to which website performed)? I have no assembly understanding it looks like the matrix for me...
How can I securly use NSStrings which cannot be read out with strings or read in assembly? I know one can do obfuscation of strings - but this is still not secure, isn't it?
This is a problem that people have been chasing for years, and any sufficiently-motivated person with skills will be able to find ways to find out whatever information you don't want them to find out, if that information is ever stored on a device.
Without jailbreaking, it's possible to disassemble apps by using the purchased or downloaded binary. This is static inspection and is facilitated with standard disassembly tools. Although you need to have a tool which is good enough to add symbols from the linker and understand method calls sufficiently to be able to tease out what's going on. If you want to get a feel for how this works, check out hopper, it's a really good disassembly/reverse-engineering tool.
Specifically to your secure log in question, you have a bigger problem if you have a motivated attacker: system-based man-in-the-middle attacks. In this case, the attacker can shim out the networking code used by your system and see anything which is sent via standard networking. Therefore, you can't depend on being able to send any form of unencrypted data into a "secure" pipe at the OS or library level and expect it not to be seen. At a minimum you'll need to encrypt before getting the data into the pipe (i.e. you can't depend on sending any plain text to standard SSL libraries). You can compile your own set of SSL libraries and link them directly in to your App, which means you don't get any system performance and security enhancements over time, but you can manually upgrade your SSL libraries as necessary. You could also create your own encryption, but that's fraught with potential issues, since motivated hackers might find it easier to attack your wire protocol at that point (publicly-tested protocols like SSL are usually more secure than what you can throw together yourself, unless you are a particularly gifted developer with years of security/encryption experience).
However, all of this assumes that your attacker is sufficiently motivated. If you remove the low-hanging fruit, you may be able to prevent a casual hacker from making a simple attempt at figuring out your system. Some things to avoid:
storing plain-text encryption keys for either side of the encryption
storing keys in specifically named resources (a file named serverkey.text or a key stored in a plist with a name which contains key are both classics)
avoid simple passwords wherever possible
But, most important is creating systems where the keys (if any) stored in the application themselves are useless without information the user has to enter themselves (directly, or indirectly through systems such as OAUTH). The server should not trust the client for any important operation without having had some interaction with a user who can be trusted.
Apple's Keychain provides a good place to store authentication tokens, such as the ones retrieved during an OAUTH sequence. The API is a bit hard to work with, but the system is solid.
In the end, the problem is that no matter what you do, you're just upping the ante on the amount of work that it takes to defeat your measures. The attacker gets to control all of the important parts of the equation, so they will eventually defeat anything on the device. You are going to need to decide how much effort to put into securing the client, vs securing the server and monitoring for abuse. Since the attacker holds all of the cards on the device, your better approach is going to be methods that can be implemented on the server to enhance your goals.
I'm working on a web application and we are getting ready to launch it. Because it will hold sensitive data for users, I want this to be as secure as possible. Here is a list of what we are currently doing...
Running the app on Heroku (Ruby on Rails)
Site is encrypted with 256 SSL (with forced SSL turned on)
Cookies are encrypted and we pass the Firesheep test
Their password and everything in the database is one way encrypted.. so even if someone got access to the database it would be useless.
We do not store any keys or passwords openly in the source code but rather use Config Vars
Other than that what else should/could we be doing. We are considering McAfee's site scan but they quoted us $2,500 a year. I'm not sure it's worth it.
Does anyone have any suggestions at all?
Make sure to read the OWASP Top 10. Also $2,500 is a rip off, Sitewatch is free. You should also consider running a Web Application Firewall like mod_security, but keep in mind this will cause problems for testing tools like McAfee or Sitewatch. You should configure mod_security to allow specific ip addresses. Or test your application before enabling the WAF.
After ruling out the usual suspects (XSS, SQL injection, mass assignment, etc), client side is where most problems come from, and this is often overlooked. I don't know what your site is about, but things like telling your users that they shouldn't follow links on emails they did not explicitly request usually delivers highest bang-for-the-buck.
Best regards,
-- J. Fernandes
I'd recommend checking out the OWASP Top 10: http://owasptop10.googlecode.com/files/OWASP%20Top%2010%20-%202010.pdf
The OWASP Top Ten provides a powerful awareness document for web application security. The OWASP Top Ten represents a broad consensus about what the most critical web application security flaws are. Project members include a variety of security experts from around the world who have shared their expertise to produce this list.
To verify your SSL configuration, you can try https://www.ssllabs.com/ssldb/index.html.
If you're curious about the sheer variety of attacks, check out Jeremiah Grossman's post titled Top Ten Web Hacking Techniques of 2010 and scroll down until you see "The Complete List".
If you want to fire off a few web app vulnerability scans tools to catch the low hanging fruit you can try:
skipfish: http://code.google.com/p/skipfish/ (free)
netsparker community: http://www.mavitunasecurity.com/communityedition/ (free)
look here for more https://security.stackexchange.com/questions/32/what-tools-are-available-to-assess-the-security-of-a-web-application/
If you're really concerned about security then adopting a secure development plan and working with someone trained in app security would obviously boost your confidence things are being done right.
Regarding development, you may like the ideas presented in Microsoft's simplified SDL:
"The Security Development Lifecycle (SDL) is a security assurance process that is focused on software development."
"The process outlined in this paper sets a minimum threshold for SDL compliance. That said, organizations aren’t uniform – development teams should apply the SDL in a way that is suitable to the human talent and resources available, but doesn’t compromise organizational security goals."
Also it is important to note automated vulnerability scan tools fail to identify most logical vulnerabilities so don't rely solely on automated tools. For example (taken from OWASP):
"Setting the quantity of a product on an e-commerce site as a negative number may result in funds being credited to the attacker. The countermeasure to this problem is to implement stronger data validation, as the application permits negative numbers to be entered in the quantity field of the shopping cart."
Human intelligence is key to spot logical issues.
Security is also all about maintenance. Assigning someone or a team the responsibility to astutely play continuous defense is important.
Note: Encrypting the passwords doesn't imply infallible security. Dictionary/password lists/brute force attacks work all the time to reveal weak passwords. A very common attack is to use SQL injection to dump the user table (with password hashes) then use a password cracker to discover legitimate user/password pairs.
You can find information about common Ruby on Rails application vulnerabilities and their countermeasures at the Zen Rails Security Checklist, including most of the OWASP Top 10 items.