Confidentiality, Integrity, and Availability (CIA) Triad

Learn the CIA triad: confidentiality, integrity, and availability, including security controls, threats, examples, and troubleshooting techniques for CCNA Security.

What Is the CIA Triad?

The CIA triad is a foundational information-security model used to guide security policies, risk decisions, and control selection. CIA stands for confidentiality, integrity, and availability.

These principles describe three different security goals:

Confidentiality: Prevent unauthorized people or systems from viewing information.
Integrity: Keep information accurate, complete, trustworthy, and changed only in authorized ways.
Availability: Ensure authorized users can access information, systems, and network services when needed.

Security decisions usually require balancing all three objectives. For example, encrypting every system and severely restricting access may improve confidentiality, but excessive restrictions could prevent legitimate users from completing their work. Similarly, keeping many redundant systems available may increase availability but also create more systems that must be secured. Effective security controls support the required balance for the organization and the data involved.

CIA Triad Principles at a Glance

Principle — Primary security goal — Typical threats or failure sources — Representative controls

Confidentiality — Prevent unauthorized disclosure — Stolen credentials, excessive permissions, eavesdropping, lost devices — Authentication, authorization, ACLs, encryption, data classification

Integrity — Preserve accurate and trustworthy information — Malware, unauthorized modification, human error, crashes, hardware failure, EMP — Hashing, access controls, backups, redundant systems, encryption where appropriate

Availability — Keep authorized services and resources accessible — Hardware failure, software faults, outages, DDoS attacks, capacity limits — Maintenance, patching, redundancy, failover, RAID, high-availability clusters, DDoS defenses

Confidentiality

Confidentiality means preventing unauthorized access to or disclosure of sensitive information. An authorized user has an approved business or operational reason to access data. An unauthorized user does not have that approval, even if the user can technically reach the system.

For example, an employee-records server may permit human-resources staff to view employee information while denying access to other departments. The goal is not to prevent all access; it is to permit the right access and prevent the wrong access.

Controls That Support Confidentiality

User identities: Unique usernames connect actions to specific users and help prevent shared-account problems.
Passwords: Strong, protected passwords make it more difficult for an attacker to use a stolen or guessed identity.
Authentication: Authentication verifies a user's or system's identity. A password, certificate, token, or biometric factor can be used as part of authentication.
Authorization: Authorization determines what an authenticated identity is permitted to read, change, or operate.
Access control lists (ACLs): An ACL is a rule set that permits or denies specified traffic or access. ACLs can restrict which users, hosts, networks, or services can reach a resource.
Encryption: Encryption transforms readable data into an unreadable form. A required key is used to recover the original data. Encryption protects data if stored media or network traffic is exposed.

Authentication answers, “Who are you?” Authorization answers, “What are you allowed to do?” Confusing these two functions can result in a user being correctly identified but granted excessive access.

Data Classification

Data classification is the process of categorizing information according to its sensitivity and the harm that could result from unintended disclosure. Typical categories might include public, internal, confidential, and highly restricted information, although the names vary by organization.

Classification should influence handling requirements. Public information may need ordinary access controls, while confidential employee records may require limited permissions, strong authentication, encryption, logging, and carefully controlled storage and transmission. The more serious the potential impact of disclosure, the stronger and more layered the confidentiality controls should be.

Example: Sensitive Employee Records

To restrict access to an employee-records server, an organization can assign unique usernames, require strong passwords and authentication, authorize only approved personnel, apply ACLs to the server and network, and encrypt the records both when stored and when transmitted. These controls reduce the chance that an unauthorized person can view the records, including a person who obtains a copy of the storage or intercepts network traffic.

Integrity

Integrity is the assurance that information remains accurate, complete, trustworthy, and suitable for its intended purpose. A recipient should receive the information that the creator or sender intended to provide.

Not every change is an integrity failure. An approved administrator updating a correct record is an authorized modification. An attacker changing a payment amount, a user accidentally overwriting a configuration, or a system corrupting a database is an unauthorized or accidental change that threatens integrity.

Integrity applies to both data at rest and data in transit. Data at rest is stored on a server, device, database, or other medium. Data in transit is moving across a network or between systems.

Hashing and Change Detection

Hashing produces a fixed-length value from data. If the data changes, recalculating the hash will normally produce a different value. Comparing a calculated hash with a trusted expected hash can therefore reveal corruption or unauthorized modification.

For example, after downloading a network-device software image, an administrator can calculate its hash and compare it with the value published through an approved distribution process. A match provides evidence that the file has not changed since the trusted hash was created. A mismatch does not identify the cause by itself; the file may have been corrupted during transfer or altered by an attacker.

Sources of Integrity Failure

Malicious alteration by an attacker or malware
Human error, such as editing the wrong file or entering incorrect data
Operating-system or application crashes
Hardware failure, including failed storage devices
Environmental events, including an electromagnetic pulse (EMP) that disrupts or damages electronic equipment and data systems

Backups and redundant systems are recovery-oriented controls. They may not prevent the original corruption, but they help preserve or restore a known-good version of data. Recovery procedures should verify that the backup itself is usable and accurate.

Example: Verifying a Software Image

When a downloaded network-device image has a hash that differs from the expected value, do not trust or deploy the file. Download it again from an approved source, recalculate the hash, and compare the values. If the mismatch persists, investigate possible transfer corruption or unauthorized modification.

Availability

Availability means ensuring that authorized users can access required information, systems, and network resources when needed. A system can be confidential and accurate but still fail its security requirements if legitimate users cannot reach it during normal operations.

Practices That Improve Availability

Hardware maintenance: Replacing aging or failing components reduces unexpected outages.
Software patching: Updates can correct defects and vulnerabilities that cause crashes or service disruption.
Network optimization: Correct capacity planning, efficient routing, and removal of bottlenecks help services respond to legitimate demand.
Redundancy: Duplicate components, links, power supplies, or systems reduce single points of failure.
Failover: Operations transfer automatically or manually to a standby component or system after a failure.
RAID: This disk-organization approach can provide storage redundancy or fault tolerance, depending on the RAID level. RAID does not replace backups.
High-availability clusters: Multiple systems work together so that a service can continue when an individual node fails.

Availability can be affected by accidental failures, such as a failed disk or software crash, and by deliberate attacks. A distributed denial-of-service (DDoS) attack uses many sources to overwhelm a target's resources or network connectivity. The target may become slow or unreachable even though its systems have not been physically damaged.

Dedicated security or network devices and services can help defend against malicious traffic. Traffic-filtering systems, DDoS mitigation devices or services, redundant capacity, and upstream protection can help preserve access for legitimate users.

Example: Recovering After a Server Crash

If authorized users cannot reach an application after a server or storage failure, identify the failed component or service first. Use failover or redundant infrastructure if it is available, verify RAID and backup status, and restore verified data from a backup if corruption or loss occurred. This response addresses availability while also checking for an integrity problem.

Applying Controls to CIA Objectives

A control can support more than one CIA objective. The table identifies the control's primary purpose, but the effects can overlap.

Control or practice — Primary CIA objective — How it helps — Example use

Usernames and passwords — Confidentiality — Restrict access to identified users — Require each administrator to use an individual account.

ACLs — Confidentiality — Permit or deny defined access or traffic — Allow only an application server to reach a sensitive database.

Encryption — Confidentiality; can also support integrity when used in an authenticated protection design — Makes exposed stored or transmitted data unreadable and may detect protected-message changes — Encrypt employee records at rest and across a network.

Hashing — Integrity — Provides a value for detecting changes — Compare the expected and calculated hashes of a downloaded image.

Backups — Integrity and availability — Preserve a known-good copy and support recovery — Restore a database after corruption caused by a crash.

Redundancy — Availability — Removes or reduces single points of failure — Provide two network paths or duplicate critical servers.

Failover — Availability — Moves service operation to a standby system — Switch to a secondary server when the primary fails.

RAID — Availability and storage resilience — Allows continued operation or recovery after certain disk failures, depending on the level — Keep a service operating after one protected disk fails.

High-availability clusters — Availability — Let another node provide the service when one node fails — Maintain application service during node maintenance or failure.

Patching and maintenance — Availability and security — Reduce faults and vulnerabilities that can cause outages or compromise — Apply tested updates and replace failing hardware.

DDoS defenses — Availability — Filter, absorb, or divert malicious traffic — Keep a public application reachable during a traffic flood.

Example Events and CIA Impact

Event — Affected CIA principle or principles — Why it matters — Mitigation approach

Unauthorized file access — Confidentiality — Sensitive information is disclosed to an unapproved user — Review authentication, authorization, ACLs, encryption, and data classification.

Altered or corrupted file — Integrity — Users may rely on inaccurate or incomplete information — Use hash comparison, access controls, backups, and verified recovery.

Server crash — Availability and potentially integrity — The service may be unreachable and active data may be damaged — Use maintenance, failover, redundancy, and verified backups.

Storage-device failure — Availability and potentially integrity — Data or the service may be lost or inaccessible — Use suitable RAID, replacement procedures, redundancy, and backups.

EMP-related disruption — Integrity and availability — Electronic equipment may be disrupted or damaged, affecting stored data and services — Use resilience planning, protected facilities where appropriate, redundancy, and recovery copies.

DDoS attack — Availability — Excessive malicious traffic can prevent legitimate access — Use traffic filtering, dedicated DDoS protection, upstream mitigation, and redundant capacity.

Troubleshooting CIA Failures

Hash Mismatch After a Download

Symptom: A file hash calculated after download does not match the expected hash.

Do not trust or deploy the file.
Download it again from an approved source.
Recalculate the hash and compare it with the expected value.
If the mismatch persists, investigate corruption or unauthorized modification.

The primary concern is integrity.

Authorized Users Cannot Reach an Application

Symptom: Users lose access after a server or storage failure.

Identify the failed component or service.
Use failover or redundant infrastructure where available.
Verify RAID and backup status.
Restore verified data from backup if corruption or loss occurred.

The primary concern is availability, with a possible integrity impact if data was damaged.

Service Unreachable During Heavy Inbound Traffic

Symptom: A service becomes unreachable during an unusually large volume of inbound traffic.

Assess the traffic volume and whether sources are distributed.
Determine whether the event resembles a DDoS attack.
Apply available traffic filtering, mitigation, or upstream protection.
Use redundant capacity and incident procedures to preserve legitimate access.

The primary concern is availability.

Unapproved User Can Read Sensitive Information

Symptom: An unapproved user can view sensitive data.

Review authentication and authorization decisions.
Inspect ACLs and account permissions.
Remove inappropriate access.
Assess whether the exposed data requires stronger encryption or classification-based controls.

The primary concern is confidentiality.

Key Terms

Authentication: Verification of a user's or system's identity.
Authorization: Determination of what an authenticated identity may access or do.
ACL: A rule set that permits or denies specified traffic or access.
Data at rest: Data stored on a device, server, database, or other medium.
Data in transit: Data moving across a network or between systems.
Redundancy: Duplicate components or paths used to reduce single points of failure.
Failover: Transfer of operations to a standby system after a failure.
DDoS: An attack in which many sources overwhelm a target's resources or connectivity.
EMP: An electromagnetic event that can disrupt or damage electronic equipment and data systems.

Exam-Relevant Summary

Confidentiality prevents unauthorized disclosure.
Integrity ensures data is accurate, complete, trustworthy, and changed only in authorized ways.
Availability ensures authorized users can access required resources when needed.
Authentication verifies identity; authorization determines permissions.
Encryption primarily protects confidentiality, while hashing commonly detects data changes.
Backups help restore integrity and availability but do not by themselves prevent every failure.
Redundancy, failover, RAID, and high-availability clusters primarily reduce downtime.
DDoS attacks primarily threaten availability.
A single control can support multiple CIA objectives, so evaluate controls in context rather than assigning each control to only one category.

For related foundations, review basic security terms and AAA concepts. For network-device protection, see control-plane protection, data-plane protection, and secure IOS images and files.