Patch Management Guide
This guide explains the patching requirements for Ministry of Justice (MoJ) systems once a vulnerability has been identified.
This guide is a sub-page to the Vulnerability Scanning and Patch Management Guide
The intent is to avoid compromise of MoJ systems because of vulnerabilities.
Parent topic: Vulnerability Scanning and Patch Management Guide
Related information
Patching of MoJ systems and equipment
This guidance must be followed for all systems and services developed or procured by the MoJ. It applies to all asset types including, but not limited to:
- Internet facing websites and digital services.
- End user client devices, such as Desktop PCs, laptops, tablets, and mobile phones.
- Infrastructure devices, such as networking equipment, servers, and printers.
- Applications.
- Internet-of-Things (IoT) devices.
In general, there are three options for patching:
- The problem is serious or urgent, and must be mitigated as soon as possible.
- The problem is important but not urgent; mitigation can wait until the next scheduled patching cycle.
- The problem does not require mitigation in advance of changes introduced as part of normal system upgrades.
Note: The nature of the patching cycle will depend on what is agreed during development, deployment, and subsequent maintenance of the system or service.
Patching is the application of a vendor-supplied security patch. It can also refer to other ways of achieving the same goal. Examples include:
- Virtual patches.
- Removal of vulnerable services or functionality.
- Disabling and preventing access.
Patching might include recompiling applications to incorporate security updates. The updates might be in third party libraries or other code.
Always apply patches as soon as possible. Where this guidance mentions a time limit, you should apply patches no later than the time given. Some important or sensitive systems might need more urgent patching. For example, a system might need you to apply ‘critical’ or ‘high risk’ patches within 7 days.
Where patching or other mitigation is required, it must be applied in compliance with the Patching Schedule in this guide.
Operating systems and applications installed on systems must be:
- Licensed and supported.
- Removed from devices when no longer licensed or supported.
- Patched as soon as possible.
- Patched within no more than 14 days of an update being released, where the fix is for a ‘critical’ or ‘high risk’ vulnerability.
To ensure that patches are implemented on systems, you must either:
- Enable and use any vendor-provided automatic patch deployment mechanisms for the system.
- If automatic patch deployment is not available, apply patches manually according to the schedule outlined in this guide.
If a system or service, or a component it depends on, can no longer be licensed or supported, it should be reviewed within the timescale of the vulnerability scanning lifecycle, to determine what action to take. If the required license or support cannot be obtained, the system or service should be replaced by an alternative, or removed. If the system or service cannot be removed, then the issue should be raised through the patching exemptions process outlined in the Patching Exemptions section in this guide.
In summary:
- It must be possible to patch or mitigate an MoJ system or service. A clear, documented process must exist explaining how to patch or mitigate.
- Wherever possible, patching should be automated, or at least have minimum possible dependency on manual intervention.
- If a patch is not available, or cannot be deployed, then a suitable risk mitigation might be acceptable.
- Patching or mitigating a system or service might impact or be affected by other systemic components. These must be identified and addressed as part of the patch or mitigation process.
MoJ Digital services
For systems or services developed by the MoJ, it must be possible to patch or mitigate in order to address any vulnerabilities. To ensure this is possible:
- The Beta development stage must include a mechanism or process for a new or updated service to track and apply patches.
- Sufficient logs must be available from the new or updated service, so that security problems can be tracked from identification through to rectification.
- The patching process must also describe how to triage and action any problems.
Patching Schedule
The following Patching Schedule defines the indicative severity ratings and consequent timescales. All vulnerabilities must be remediated or patched in line with this schedule. By agreement and formal approval, alternative timescales for system patching, on a case-by-case basis, can be operated.
Note: The default is for patches to be applied as soon as possible. You should not normally delay patching because of concerns about possible issues with the patches themselves.
Patches and updates for security related devices must be treated as High Risk (3) at least, and implemented in accordance with this rating.
For ratings of High Risk (3) or Critical (4), the Risk Advisor team (contact through email) must evaluate the probability and impact, and use this to guide a ‘tolerance’ period, at the end of which a patch must be applied.
Where the rating is Medium (2) or lower, the patch can be deferred to the next scheduled maintenance or patching activity.
This schedule outline is considered a baseline. Some systems might require different patching schedules. These different schedules must be identified in the system’s Information Risk Assessment Report (IRAR).
An IRAR is normally completed by Security Architects and Risk Assessors, in conversation with the system architects, designers and developers. The IRAR document must also be agreed with the Business Continuity Team. For more information regarding IRARs, and how to create and maintain them, contact the Security team.
Rating (Severity) | Infrastructure Devices; Server Applications; Digital Services | End User Client Devices | Web Check Reporting |
---|---|---|---|
Critical (4) | 3-7 days after vulnerability alert released. | 14 days after vulnerability alert released. | Urgent: Serious configuration problems that you should fix without delay and no later than 28 days after the vulnerability alert is released. |
High Risk (3) | 3-7 days after vulnerability alert released. | 14 days after vulnerability alert released. | Advisory: Configuration problems that leave the site vulnerable. Patches should be implemented no later than 28 days after the vulnerability alert is released. |
Medium (2) | 28 days after vulnerability alert released. | 28 days after vulnerability alert released. | Informational: Configurations that you could optimise, or information that you may find useful. |
Low (1) | Next scheduled system upgrade (not to exceed 90 days). | Next scheduled system upgrade (not to exceed 90 days). | N/A |
Positive (0) | N/A | N/A | Positive: Site configurations that conform to best practices. |
Patch management processes
There are two patching and change management approaches in the MoJ.
Infrastructure and services provided by a Managed Service Provider (MSP)
Where services and infrastructure are provided by MSPs, the vendor is normally responsible for developing and implementing patches to identified vulnerabilities. Patches or a workaround must be provided by vendors to ensure the MoJ is able to meet the schedule for implementing patches.
If this is not possible, vendors must provide a firm statement that the patch or workaround cannot be made available within the timescale mandated for addressing the vulnerability. The vulnerability alert must then be escalated to the Risk Advisor team (contact through email) for help with acceptance, transfer, mitigation or avoidance.
Any patches to be deployed must go through the normal change management and approval process, and the changes recorded in the Service Management Tool.
Services and applications developed by MoJ in-house project teams
Where services and applications are developed by MoJ in-house project teams, patching and change management is addressed on a project-by-project basis. Changes are identified through awareness channels and scanning activities. These identify operational and security issues. A change management ticket must be created, detailing the change required. The project manager follows the change management process to determine how and when to implement the change, based on the security risk rating.
The patch review and approval is normally managed within the project team. If assistance is required, contact the Security Team.
If changes are urgent because a major security risk has been identified, the product, system, or service owner should ask a competent developer to investigate, and if possible create and implement a patch quickly. If the issue is more complex, Technical Architects, Security Architects and the Security Team might need to assist in the development of appropriate remediation plans.
Patches must be implemented according to the schedule in this guide. If this is not possible, the project team must provide an indication that the patch or workaround cannot be implemented within the timescale mandated for addressing the vulnerability. This delay must be escalated to the Risk Advisor team (contact through email) for help with acceptance, transfer, mitigation or avoidance.
Removal of equipment
If a system or service vulnerability cannot be patched or mitigated, it might be necessary to remove that system or service.
Before the removal of any system or service, a fresh Business Impact Assessment must be conducted and the business process owner consulted. The removal of a system or service is likely to come under the emergency and major change process.
Patching exemptions
In exceptional cases where patching of systems is not possible, other mitigations (such as logical separation) must be identified and evaluated for efficacy prior to enablement. The circumstances must be discussed with the affected Information Asset Owners (IAO) and System Owners. If the IAOs agree with the deviation, System Owners must request formal approval by the Senior Information Risk Owner (SIRO) for the exemption. Approval must be sought and obtained within a comparable timescale to applying a patch. If a critical patch cannot be applied, the approval to be exempt must be obtained within the same number of days allowed for applying a critical patch.
Contact details
Contact the Security Team for advice on risk, scanning and patching, to report a vulnerability alert, or to add a URL to the MoJ Web Check system: security@justice.gov.uk.
Feedback
If you have any questions or comments about this guidance, such as suggestions for improvements, please contact: itpolicycontent@digital.justice.gov.uk.