Niklas Bjorkman wrote: Firstly I agree with your conclusion. NewSQL takes the best of the traditional databases and NoSQL databases to combine the benefits of both worlds. I do not agree that NewSQL vendors focus on giving scale-out features to transactional data. The NewSQL market is focusing on giving true ACID support combined with extreme performance, stepping away from the traditional relational structures in databases. A lot of developers appreciate the ease of accessing data using SQL and I think we will see more and more databases supporting standard SQL.
As you said - NewSQL databases often maintain the...
In many cases, the end of the year gives you time to step back and take stock of the last 12 months. This is when many of us take a hard look at what worked and what did not, complete performance reviews, and formulate plans for the coming year. For me, it is all of those things plus a time when I u...
If you look at some of the headline-making breaches of the past few years, they all occurred at large companies with highly dynamic and complex computing environments. Securing these environments is impossible to do without automation, which is why so much of the innovation in IT security in recent years has been focused on automating security management.
Network security is one area where systems have become too complex to manage manually. Let's take firewalls as a case in point. A single firewall can have hundreds or thousands of rules, each made up of three components: source, destination and service. Next-generation firewalls add at least two additional fields - users and applications.
Larger companies have hundreds of firewalls, usually from multiple vendors, in multiple geographies, managed by different people. That's just for starters - any number of factors can exponentially increase the degree of firewall complexity. While rare, in extreme situations a particularly bloated or neglected rule base - or even a simple typo while configuring a rule - if left untended, can result in a situation where a firewall can introduce more risk than it prevents.
The only way to get your brain around the state of your firewalls is to audit them. While manual audits can be painful (not to mention error prone), it may be the best way to fast-track proper firewall management. Regardless of whether an audit is manual or automated, here are some important metrics to look for:
Number of shadowed or redundant rules: Shadowed rules refer to rules that are masked, completely or partially, by other rules that are either placed higher up in the rule base. Shadowed rules are very common because administrators add new firewall rules all the time but for a variety of reasons - fear of causing an outage, or not knowing why a rule was added in the first place - rarely delete them. A rule base filled with shadowed rules is not only inefficient, it puts a much greater strain on the firewall then is necessary, which can lead to performance issues.
Number of unused rules: Unused rules will appear rarely, if at all, in firewall logs because they aren't being used for legitimate traffic. Unused rules can lead to serious exposures, such as allowing access to a server that is no longer being used and, as a result, exposing a service likely not properly patched. Looking for unused rules manually can be an extremely slow and tedious process, which is why admins hate to do it. However, automated solutions can make auditing for unused rules both manageable and simple.
Number of unused objects: An object is a component of a rule, and a single field of a rule (i.e., source, destination or service) can have multiple objects - such as a business unit having access to multiple destinations and/or services. Not only do unused objects appear much more frequently than unused rules, they are that much harder to find manually. Cleaning up unused objects can significantly tighten up a rule base and often lead to improved performance.
Number of rules with permissive services: The most common examples of this are rules with "ANY " in the service field, but in general, permissive services give more access then is needed to the destination by allowing additional services (which are often applications), which can lead to unauthorized use, allow the service to be a springboard to other parts of the network, or leave it exposed to malicious activity.
Number of rules with risky services(such as telnet, ftp, snmp, pop, etc.) in general or between zones (i.e., between Internal, DMZ, External, or between development and production networks): Risky services are deemed risky because they usually allow credentials to be passed in plain text, often contain sensitive info or enable access to sensitive systems. Any service that exposes sensitive data or allows for shell access should be tightly monitored and controlled.
Number of expired rules: Any rule that was created on a temporary basis and has clearly expired is just taking up space and does not need to remain in the rule base. If there is no documentation as to when or why the rule expired, check the firewall logs for its "hit count" (or usage, in firewall management-speak).
Number of unauthorized changes: These are rules that are not associated with a specific change ticket. In order to ensure all requests are properly handled, all requests, from initial request to final implementation, and should be managed via a ticketing system. If a change causes a problem and no one has any clue why the change was made, who made it, whether it was assessed for compliance and/or security risk, who approved, etc., it results in a huge waste of resources, and is a clear indicator that firewall management processes are inefficient.
Percentage of changes made outside of authorized change windows: Outages in IT are usually caused by the work IT does. In order to minimize business disruption, change windows are usually set during times when the least amount of people are on the network - for example, firewall admins like to implement rule changes late at night, usually on the weekends. That way, if there is an outage it is much easier to track which change caused it and how. If the majority of changes are made outside of normal change windows, it is usually because admins are spending too much time putting out fires, which indicates the firewall management processes are in need of an overhaul.
Number of rules with no documentation: While the comments section of a firewall rule has text limits that inhibit proper documentation, all change tickets have a comments section, which can be used to provide a business justification for the rule. Some people use spreadsheets to capture this information although using a spreadsheet as a central repository for change information leads to the same issues as with other manual processes. Without proper documentation there is no way to know why a rule was implemented and if it is still necessary.
Number of rules with no logging: Proper firewall management is impossible without leveraging the data found in firewall logs. Similar to other areas of IT, there was a resistance to turning on logging because it would cause performance issues. However, firewall (and firewall management) technology has evolved to the point where logging no longer impacts performance. If there are performance issues with your firewalls, than something else in your environment is not optimal. Determining rule and object usage (numbers two and three in this list) are impossible to do without logging, as is proper forensics and troubleshooting. The advent of automated tools for firewall management makes scanning logs for relevant data a highly manageable endeavor.
Plenty of lip service has been given to the credo that good IT is a function of people, process and technology. Given the fast pace of modern business, breakdowns are bound to happen, and unless a company decides to not use the Internet, they will always be facing a certain level of exposure. While there will never be silver bullet for security, the good news is that proper attention to the metrics above will keep your firewalls optimized so that they remain one of your strongest and most reliable security assets and not a potential liability.
About Michael Hamelin Michael Hamelin is Chief Security Architect at Tufin Technologies where he identifies and champions the security standards and processes for Tufin. Bringing more than 16 years of security domain expertise to Tufin, Hamelin has deep hands-on technical knowledge in security architecture, penetration testing, intrusion detection, and anomalous detection of rogue traffic. He has authored numerous courses in information security and worked as a consultant, security analyst, forensics lead, and security practice manager. He is also a featured security speaker around the world widely regarded as a leading technical thinker in information security.
Hamelin previously held technical leadership positions at VeriSign, Cox Communications, and Resilience. Prior to joining Tufin he was the Principal Network and Security Architect for ChoicePoint, a LexisNexis Company. Hamelin received BS degrees in Chemistry and Physics from Norwich University, and did his graduate work at Texas A&M University.
SYS-CON Events announced today that MetraTech Corp., the leading provider of agreements-based billing™, commerce and compensation solutions, has been named “Bronze Sponsor” of SYS-CON's 12th International Cloud Expo, which will take place on June 10–13, 2013, at the Javits Center in Ne...
Cloud computing is more than a buzz-phrase it’s a transformative IT paradigm shift. The emphasis in the cloud is on elasticity, scalability, agility and open. Not just open standards but open APIs and open source. The delivery of software is also going through a paradigm shift. Open so...
Service Component Architecture (shortly referred as SCA) is a technology for creating services from components. SCA is a set of OASIS standards and part of it is developed with the collaboration of vendors from open source community, referred as “OSOA” Open SOA. SCA helps to build syst...
In an ideal developer/systems administrator’s world, most applications would deploy seamlessly to multiple platforms and scale elastically with minimal effort bringing the unprecedented agility of the cloud within immediate reach of developer teams and IT organizations.
OpenStack, a ...
Storage and Archive offerings are now exploding on the market. From end-user mobile devices to company tactical level, the cloud has become a black hole for every kind of data. But what are the risks, and what are the real needs?
In his session at the 12th International Cloud Expo, A...
Companies around the world are collecting massive amounts of data everyday that’s sitting around and not being utilized. Take for example the fact that companies collect demographic and location-based data via mobile devices all the time, but have to figure out how to monetize that dat...