Compliance

What is GDPR? GDPR or General Data Protection Regulation is a regulation to protect personal data from citizens of the European Union (EU). When speaking about stored data, it includes the handling of data at any given time, from the initial creation of the data, until the final deletion of it. One of the important parts is the right to ‘know’. That means that individuals can ask what data is stored about them.

The standard itself is very detailed. Still, it sometimes unclear on what specifically to implement and when. This guide will help with translating the PCI standard to technical security controls on Linux systems. This document has the goal to help you further secure your network and pass the PCI DSS audit. It is important to note that this guide is a set of generic tips. Your IT environment might require additional security measures.

If you are involved with Linux security, you might already have come across SOx compliance. Usually from a sysadmins point of view, work doesn’t get easier due to these compliance requirements. Still there are some lessons we can learn, followed by a great documentary to watch about Enron. The Sabanes-Oxley Act, SOx for short, now applies to all companies who like to have their stock exchanged at the New York stock exchange (NYSE).

Some areas are within the PCI DSS standard are definitely not directly clear when reading the description. Section 10.2.7 is one of them. It talks about the creation and deletion of system-level objects and specifically the ability to log them. System-level objects? The guidance in 10.2.7 speaks about malware and mentions database related items. That does not make auditing very obvious, as malware usually targets binaries. Therefore we have to look first what a system-level object is.

PCI describes in control 10.2.4 to monitor for “invalid logical access attempts”. Another way of saying to monitor attempts which are not allowed, like accessing a file you are not supposed to. Another indication might be brute force attempts to log in, which result in several failed logins. To monitor for invalid access attempts, we can use the Linux audit framework. This framework has been created and maintained by Red Hat over the years.

Companies who need to comply with the PCI DSS standard need to log all actions which are executed by the root user or those accounts with similar administrative privileges. 10.2.2 Verify all actions taken by any individual with root or administrative privileges are logged. The Linux kernel allows the monitoring of executed commands. This monitoring and logging can be done with the Linux audit framework. Using this framework, we can monitor the right system calls and create an audit trail.

A.1.2.c Verify that an entity’s users do not have write access to shared system binaries Shared system binaries should be protected, as they form the basis of your system. PCI compliance (A.1.2.c) demands that users do not have write access to shared systems binaries. The only exception is of course the root user, so software upgrades are still possible. Paths for system binaries Depending on the distribution used there are several directories which have shared system binaries.

A.1.2.a Verify the user ID of any application process is not a privileged user (root/admin). For Unix and Linux based systems, processes should run as a non-privileged user where possible. However to be able to start, a process is usually started with root permissions (uid 0). This is required to open the required sockets (e.g. bind to port 80). After the initial start, the process drops its privileges by switching to another user.

A.1.2.d Verify that viewing of log entries is restricted to the owning entity. To limit exposure to information, PCI DSS requires access of logging to only the entity owning that log file. In other words, we have to search for those entries which can be seen by others. Search related log files By default, most log files on Linux based systems will be stored in /var/log. We can do a quick check for any files which are world readable, by using find.

Determining the level of Linux server security can only by measuring the actual implemented security safeguards. This process is called auditing and focuses on comparing common security measures with the ones implemented. While there is almost no system with all possible safeguards implemented, we still can determine how well (or badly) the system is protected. Security is about finding the weakest link(s) and associate risk with each weakness. Depending on the role of the system, sensitivity of data and possible threats, we can then select what security safeguards are appropriate.