The flaw was discovered by researchers from mobile security firm Appthority and impacts apps deployed on iOS devices through mobile device management (MDM) or enterprise mobility management (EEM) products. These products allow administrators to automatically push applications, configuration settings and data access rules to enterprise mobile devices.
Before a new iOS device is brought inside the network of a company that uses a mobile management system, an MDM account is created for it and a client application is installed. The MDM client is used to install corporate apps and to enforce access policies for corporate data and email.
According to Appthority, when corporate apps are deployed in this manner, their sensitive configuration settings, which can include internal server URLs, usernames, passwords and access tokens, are stored in a "world readable" location.
Because of this, any other application installed on the same device can read that information, violating the standard iOS application sandbox that's supposed to prevent apps from reading each others' data.
Attackers could create a rogue application designed to read sensitive information from apps that use managed configurations and could distribute that app through the iTunes store or via email-based spear phishing attacks, the Appthority researchers said in a blog post. To increase the chance of affecting enterprise users, the attackers could design the rogue app to work as a productivity tool, they said.
It could be hard to detect the rogue app's malicious behavior because it would be just one of many legitimate applications with access to the directory where the managed configurations are stored.
The impact to enterprises depends on the kind of information that might be exposed through their managed app configurations. Appthority scanned millions of apps and found that the majority of those with vulnerable configuration data were MDM clients, corporate apps that grant access to work email and business documents and secure browsers used on internal networks.
"We also found apps used in the healthcare industry, giving doctors access to patient data and records," whose compromise could result in possible violations of the Health Insurance Portability and Accountability Act (HIPAA), the researchers said.
Almost half of the managed settings for the identified applications contained credentials like usernames, passwords and access tokens, and 67 percent contained information identifying servers, according to Appthority.
Apple fixed the vulnerability in iOS 8.4.1 that was released on Aug. 13 by hardening the sandbox profile for apps with managed preferences. However, based on previous scans, Appthority determined that up to 70 percent of iOS devices don't get updated to the latest iOS version even months after it was released.