# Indicators for secure access and delegation Establish scalable, fine-grained access controls that balance security with team autonomy. Granting explicit, temporary access based on the principle of least privilege, providing procedures for emergencies, and regularly auditing access controls to align with evolving requirements and threats. **Topics** + [ # [AG.SAD.1] Centralize and federate access with temporary credential vending ](ag.sad.1-centralize-and-federate-access-with-temporary-credential-vending.md) + [ # [AG.SAD.2] Delegate identity and access management responsibilities ](ag.sad.2-delegate-identity-and-access-management-responsibilities.md) + [ # [AG.SAD.3] Treat pipelines as production resources ](ag.sad.3-treat-pipelines-as-production-resources.md) + [ # [AG.SAD.4] Limit human access with just-in-time access ](ag.sad.4-limit-human-access-with-just-in-time-access.md) + [ # [AG.SAD.5] Implement break-glass procedures ](ag.sad.5-implement-break-glass-procedures.md) + [ # [AG.SAD.6] Conduct periodic identity and access management reviews ](ag.sad.6-conduct-periodic-identity-and-access-management-reviews.md) + [ # [AG.SAD.7] Implement rotation policies for secrets, keys, and certificates ](ag.sad.7-implement-rotation-policies-for-secrets-keys-and-certificates.md) + [ # [AG.SAD.8] Adopt a zero trust security model, shifting towards an identity-centric security perimeter ](ag.sad.8-adopt-a-zero-trust-security-model-shifting-towards-an-identity-centric-security-perimeter.md) # [AG.SAD.1] Centralize and federate access with temporary credential vending **Category:** FOUNDATIONAL Implement a centralized subsystem for federated access and temporary credential vending to maintain secure and controlled access to your environments, workloads, and resources. By implementing a federated access solution, you can leverage your existing identity systems, provide single sign-on (SSO) capabilities, and avoid the need to maintain separate user identities across multiple systems which makes scaling in a DevOps model more tenable. Centralizing identity onboarding and permission management eliminate the inefficiencies of manual processes, reduce human error, and enable scalability as your organization grows. Grant users and services fine-grained access to help ensure secure, granular control as they interact with resources and systems. By applying the least privilege principle, you can minimize the risk of unauthorized access and reduce the potential damage from compromised keys while retaining full control over access to resources and environments. To reduce the likelihood of keys being compromised, always vend short-lived, temporary credentials that are scoped for specific tasks to help ensure that privileges are granted only for the duration needed. **Related information:** + [AWS Well-Architected Cost Optimization Pillar: COST02-BP04 Implement groups and roles](https://docs.aws.amazon.com/wellarchitected/latest/cost-optimization-pillar/cost_govern_usage_groups_roles.html) + [Security best practices in IAM](https://docs.aws.amazon.com/IAM/latest/UserGuide/best-practices.html) + [AWS Well-Architected Security Pillar: SEC02-BP04 Rely on a centralized identity provider](https://docs.aws.amazon.com/wellarchitected/latest/security-pillar/sec_identities_identity_provider.html) + [IAM Identity Center](https://docs.aws.amazon.com/singlesignon/latest/userguide/what-is.html) + [What is SSO (Single-Sign-On)?](https://aws.amazon.com/what-is/sso/) + [Identity providers and federation](https://docs.aws.amazon.com/IAM/latest/UserGuide/id_roles_providers.html) # [AG.SAD.2] Delegate identity and access management responsibilities **Category:** FOUNDATIONAL Create a decentralized Identity and Access Management (IAM) responsibility model that enables individual teams to handle their own IAM tasks, such as creating roles and assigning permissions, as long as those teams operate within applied guardrails. This approach grants teams the autonomy to manage their roles and permissions essential for the applications they develop, encourages a culture of ownership and accountability, and enables your organization to scale its permission management effectively as it grows and embraces more DevOps practices. Establish a set of well-defined guardrails which limit the maximum permissions a user or role can safely have. These guardrails reduce potential security risk while creating balance between allowing teams to manage their own IAM tasks and ensuring that they do not exceed the maximum permissions set. **Related information:** + [Security best practices in IAM](https://docs.aws.amazon.com/IAM/latest/UserGuide/best-practices.html) + [Use permissions boundaries to delegate permissions management within an account](https://docs.aws.amazon.com/IAM/latest/UserGuide/best-practices.html#bp-permissions-boundaries) + [Establish permissions guardrails across multiple accounts](https://docs.aws.amazon.com/IAM/latest/UserGuide/best-practices.html#bp-permissions-guardrails) + [Blog: Delegate permission management to developers by using IAM permissions boundaries](https://aws.amazon.com/blogs/security/delegate-permission-management-to-developers-using-iam-permissions-boundaries/) # [AG.SAD.3] Treat pipelines as production resources **Category:** FOUNDATIONAL Pipelines become pivotal in every aspect of the software development lifecycle when practicing DevOps, as they become the sole method of moving code from development to production. During the process of building, testing, and deploying software, pipelines require access to all software components involved, including libraries, frameworks, repositories, modules, artifacts, and third-party dependencies. Due to this level of access and their role in deploying to potentially sensitive environments, pipelines should be recognized as integral components of your overall system and must be secured and managed to the same degree as the environments and data they interact with. The [application of least-privilege principles](https://docs.aws.amazon.com/IAM/latest/UserGuide/best-practices.html#grant-least-privilege), commonly applied to human users, should be extended to pipelines. To reduce the potential for pipelines to become a security threat, their roles and permissions should be confined to align with their precise responsibilities. Emphasizing pipeline governance and treating pipelines as first-class citizens within your security infrastructure can substantially decrease your potential attack surface and reinforce the security of your overall DevOps environment. **Related information:** + [AWS Well-Architected Security Pillar: SEC11-BP07 Regularly assess security properties of the pipelines](https://docs.aws.amazon.com/wellarchitected/latest/framework/sec_appsec_regularly_assess_security_properties_of_pipelines.html) # [AG.SAD.4] Limit human access with just-in-time access **Category:** FOUNDATIONAL As pipelines take on a more prominent role in the software development lifecycle in a DevOps model, the necessity for extensive human access to environments decreases. Human users should be granted minimal access necessary for their role, which is usually read-only access that does not allow any modifications or access to sensitive data. For experimentation which is typically hands-on and exploratory, teams should be granted access to sandbox environments which are isolated from system workloads. In some cases, where things go wrong or a process cannot yet be automated, elevated permissions might be required. To accommodate these needs without compromising security, implement a just-in-time (JIT) access control strategy where permissions are temporarily escalated for a specific duration and purpose, upon explicit request and approval. This approach maintains the principle of least privilege, allowing necessary operational functions to be performed efficiently when needed, while also ensuring that the access is revoked once the task is complete. By enforcing limited human permissions and using JIT access, you can improve your organization's security posture and reduce the risk of accidental or deliberate misuse of access rights. This restrictive and controlled model supports modern, secure DevOps practices where pipelines, treating everything as code, and automation should take precedence over manual actions. **Related information:** + [Eliminate the need for human access](https://docs.aws.amazon.com/wellarchitected/latest/financial-services-industry-lens/use-immutable-infrastructure-with-no-human-access.html) + [AWS Samples: AWS IAM Temporary Elevated Access Broker](https://github.com/aws-samples/aws-iam-temporary-elevated-access-broker) + [Blog: Managing temporary elevated access to your AWS environment](https://aws.amazon.com/blogs/security/managing-temporary-elevated-access-to-your-aws-environment/) # [AG.SAD.5] Implement break-glass procedures **Category:** FOUNDATIONAL Emergencies or unforeseen circumstances might necessitate temporary access beyond regular permissions for day-to-day work. Having break-glass procedures helps ensure that your organization can respond effectively to crises without compromising long-term security. During emergency scenarios, like the failure of the organization's identity provider, security incidents, or unavailability of key personnel, these measures provide temporary, elevated access beyond regular permissions. Implement measures that improve the resilience of your DevOps environments through the ability to respond effectively to emergencies without compromising long-term security. Create break-glass roles and users you can assume control of during emergencies that are able to bypass established controls, update guardrails, troubleshoot issues with automation tooling, or remediate security and operational issues that may occur. These break-glass roles and users should have adequate security measures, such as configuring them with hardware-based multi-factor authentication (MFA), to ensure that even in emergencies, access is tightly controlled and auditable. Establish alerts and alarms triggered by the use of these break-glass roles and users, and tie their usage closely to incident response and recovery procedures. **Related information:** + [AWS Well-Architected Security Pillar: SEC03-BP03 Establish emergency access process](https://docs.aws.amazon.com/wellarchitected/latest/security-pillar/sec_permissions_emergency_process.html) + [Break glass access](https://docs.aws.amazon.com/whitepapers/latest/organizing-your-aws-environment/break-glass-access.html) + [Amazon's approach to high-availability deployment: Dealing with the real world](https://youtu.be/bCgD2bX1LI4?t=1300) # [AG.SAD.6] Conduct periodic identity and access management reviews **Category:** FOUNDATIONAL With the distributed nature of DevOps Identity and Access Management (IAM) responsibilities, it is important to systematically review IAM roles and permissions periodically. This helps ensure that changes in roles and permissions align with the rapidly shifting needs of the organization, and that the guardrails set in place for delegation are working as intended or perhaps need to be fine-tuned. This activity aids in identifying unused or overly broad permissions, reinforcing the adherence to the principle of least privilege and reducing potential security risks. Optionally, automate the right-sizing of permissions as part of these reviews. This proactive approach not only keeps IAM policies up-to-date, but also minimizes potential avenues for unauthorized access, further strengthening your overall security posture. Automatically right sizing roles and permissions based on actual activity allows organizations to scalably enforce that the right resources are accessible to the right entities, at the right times. **Related information:** + [AWS Well-Architected Security Pillar: SEC03-BP04 Reduce permissions continuously](https://docs.aws.amazon.com/wellarchitected/latest/security-pillar/sec_permissions_continuous_reduction.html) + [Regularly review and remove unused users, roles, permissions, policies, and credentials](https://docs.aws.amazon.com/IAM/latest/UserGuide/best-practices.html#remove-credentials) + [Use IAM Access Analyzer to generate least-privilege policies based on access activity](https://docs.aws.amazon.com/IAM/latest/UserGuide/best-practices.html#bp-gen-least-privilege-policies) + [Verify public and cross-account access to resources with IAM Access Analyzer](https://docs.aws.amazon.com/IAM/latest/UserGuide/best-practices.html#bp-preview-access) + [Using AWS Identity and Access Management Access Analyzer](https://docs.aws.amazon.com/IAM/latest/UserGuide/what-is-access-analyzer.html) + [Blog: IAM Access Analyzer makes it easier to implement least privilege permissions by generating IAM policies based on access activity](https://aws.amazon.com/blogs/security/iam-access-analyzer-makes-it-easier-to-implement-least-privilege-permissions-by-generating-iam-policies-based-on-access-activity/) + [Blog: Continuous permissions rightsizing to ensure least privileges in AWS using CloudKnox and AWS Config](https://aws.amazon.com/blogs/mt/continuous-permissions-rightsizing-ensure-least-privileges-aws-using-cloudknox-aws-config/) # [AG.SAD.7] Implement rotation policies for secrets, keys, and certificates **Category:** RECOMMENDED Regular rotation of secrets, keys, and certificates is a best practice in securing access, limiting the potential damage that can occur should these security resources become compromised. In a DevOps environment, pipelines often require access to sensitive environments and workloads, making them potential targets for attacks. The routine rotation of these resources that are used by pipelines can help to significantly mitigate this risk. Establish a policy that clearly defines the lifecycle of these resources, including their creation, usage, rotation, and retirement intervals. Enforce these policies by automatically rotating secrets and keys to reduce the risk of oversights, delays, and human error. Certificates play an important role in service-to-service authentication and providing encryption for both internal and external facing workloads and environments. When managing certificates, consider not only those issued within your organization but also those imported from external sources which may not be automatically renewable. Monitoring systems that track the lifespan of these assets and alert administrators as they near expiration can contribute to this process. This approach can help prevent service disruptions caused by expired certificates and, in some cases, can trigger automated renewal procedures. **Related information:** + [Blog: How to monitor expirations of imported certificates in AWS Certificate Manager (ACM)](https://aws.amazon.com/blogs/security/how-to-monitor-expirations-of-imported-certificates-in-aws-certificate-manager-acm/) + [Rotate AWS Secrets Manager secrets - AWS Secrets Manager](https://docs.aws.amazon.com/secretsmanager/latest/userguide/rotating-secrets.html) + [Managing access keys for IAM users - AWS IAM](https://docs.aws.amazon.com/IAM/latest/UserGuide/id_credentials_access-keys.html) + [Rotating AWS KMS keys - AWS Key Management Service](https://docs.aws.amazon.com/kms/latest/developerguide/rotate-keys.html) # [AG.SAD.8] Adopt a zero trust security model, shifting towards an identity-centric security perimeter **Category:** RECOMMENDED When operating under a zero trust security model, no user or system is trusted by default. It requires all users and systems, even those inside an organization's network, to be authenticated, authorized, and continuously validated to ensure secure configurations and posture. Only after validation will they be granted access to applications and data. Zero trust is beneficial throughout the entire software development lifecycle. From the initial stages of code development as developers interact with source code repositories, through continuous integration using internal and external tools to build and test software, to the deployment and maintenance of the workloads, each user, pipeline, third-party, and service needs to be authenticated and authorized with every request. In these scenarios, zero trust enforces adherence to the principle of least privilege, ensuring that all of these independent users and systems are granted access to the right resources only when necessary. Shifting to a zero trust model is not an all-or-nothing endeavor, it is a gradual process consistent with the DevOps principles of continuous improvement. Start small by pinpointing use cases that align with your organization's unique needs and the value and sensitivity of your systems and data. This understanding will guide the selection of zero trust principles, tools, and patterns that are most beneficial for your organization. Adopting zero trust often involves rethinking identity, authentication, and other context-specific factors like user behavior and device health. Enhance existing security practices over time, improving both identity-based and network-based security measures that complement each other to create a secure perimeter where identity-centric controls can operate. AWS provides several use cases that illustrate zero trust principles: + **Signing API requests:** Every AWS API request is authenticated and authorized individually, regardless of the trustworthiness of the underlying network. + **Service-to-service interactions:** AWS services authenticate and authorize calls to each other using the same security mechanisms used by customers. + **Zero trust for internet of things (IoT):** AWS IoT extends the zero trust model to IoT devices, enabling secure communication over open networks. **Related information:** + [Zero Trust on AWS](https://aws.amazon.com/security/zero-trust/) + [Zero Trust Maturity Model](https://www.cisa.gov/sites/default/files/2023-04/zero_trust_maturity_model_v2_508.pdf) + [Amazon Verified Permissions](https://aws.amazon.com/verified-permissions/) + [AWS Verified Access](https://aws.amazon.com/verified-access)