According to the former Equifax CEO’s testimony to Congress, one of the primary causes of this now infamous data breach was the company’s failure to patch a critical vulnerability in the open source Apache Struts Web application framework. Equifax also waited a week to scan its network for apps that remained vulnerable.[1] Would you like to appear at the next Congressional hearing on patch management?
Patch management is the process of identifying, acquiring, installing, and verifying patches for
products and systems. Patches not only correct
security and functionality problems in software and firmware, but they also
introduce new, and sometimes mandatory,
capabilities into the organization’s IT
environment. It is so useful, the CERT® Coordination Center
(CERT®/CC) claims that 95
percent of all network intrusions are avoidable by using proper patch
management to keep systems up-to-date.
This nightmare true story and compelling endorsement from
CERT®/CC, however, masks the ugly operational patch management implementation
complexities. Key enterprise challenges include:
- Timing, prioritization, and testing of patches often present conflicting requirements. Competitive prioritization of IT resources, business imperative, and budget limitations often leave patching tasks on the back burner
- Technical mechanisms and requirements for applying patches may also conflict and may include:
- Software that updates itself with little or no enterprise input
- Use of a centralized management tool
- Third-party patch management applications
- Negative or unknown interactions with network access control, health check functions, and other similar technologies
- User initiated manual software updates
- User-initiated patches or version upgrades
- Typical enterprise heterogeneous environment that includes
- Unmanaged or user managed hosts
- Non-standard IT components that require vendor patching or cannot be patched
- Enterprise owned assets that typically operate on non-enterprise networks
- Smartphones, tablets, and other mobile devices
- Patching of rehydrating virtual machines
- Firmware updates
Piling up on these purely operational tasks are the change
management steps associated with:
- Maintaining current knowledge of available patches;
- Deciding what patches are appropriate for particular systems;
- Ensuring proper installation of patches;
- Testing systems after installation; and
- Documenting all procedures and any specific configurations.
This challenge can
also be significantly exacerbated in an
IT environment that blends legacy, outsourced and cloud service provider
resources. Environment heterogeneity and the sheer
volume of patches released is why any patching strategy that primarily
relies solely on manual implementation is untenable.
According to the SANS
Institute, meeting the patch management challenge requires the creation of
a patch management methodology and the automation of that methodology.[2]
The methodology itself should include:
- A detailed inventory of all hardware, operating systems, and applications that exist in the network and the creation of the process to keep the inventory up-to-date.
- A process to identify vulnerabilities in hardware, operating systems, and applications.
- Risk assessment and buy-in from management and business owners.
- A detailed procedure for testing patches before deployment.
- A detailed process for deploying patches and service packs, as well as a process for verification of deployment.
As for the automation component, it should deliver an automated, comprehensive server lifecycle approach
that can provision and configure software, update patches and implement configurations
that can improve security and compliance across physical, virtual and cloud
servers.
It should also encompass a policy-based approach with
support for all major operating systems on physical servers and leading
virtualization and cloud platforms. An ability to automate continuous
compliance checks and remediate any
security or regulatory shortcoming is also paramount. If appropriately implemented, IT Staff should be
able to manage patching via a web
interface. Having this feature increases server to admin ratio, enhances operational productivity,
accelerates audit timelines and reduces incident response latency.
A leading solution in this space is BladeLogicServer Automation by BMC. It was
specifically designed to address the dual enterprise requirements of (1) ensuring
compliance with rules and regulations and (2) software patching to reduce
security vulnerabilities. In the market
for over 10 years, it is a comprehensive server lifecycle automation solution
that helps organizations provision and configure
software, update patches and configurations to improve security and compliance
across physical, virtual and cloud servers. Advanced
capabilities include script automation, compliance tracking and the ability to stage and test patches before
committing them. The latter feature is used to copy patch bundles to the
targeted servers before maintenance windows open.The full-function suite
integrates with change management systems to facilitate change record creation.
Vulnerability management and remediation are automated by importing
vulnerability management scan data from vendors like Qualys, Tenable and Rapid
7, and mapping the vulnerabilities back to underlying patches in BladeLogic.
Secure IT operations start with the identification and
prioritization of critical
vulnerabilities paired with the capability to deliver multi-tier remediation. These reinforcing
goals are why an advance patch automation solution is a “must have” for
today’s modern enterprise.
This post is brought to you by BMC and IDG. The views and
opinions expressed herein are those of the author and do not necessarily
represent the views and opinions of BMC.
