Study Guide
Key Management, Cryptographic Hardware and Escrow
Looking after keys from creation to destruction: the key lifecycle, entropy and randomness, centralised and decentralised management and KMIP, what TPMs, HSMs and secure enclaves protect, FIPS 140-2 and 140-3, and recovering access with key escrow and M-of-N control.
Lesson 6.7 Cryptography & PKI
Strong encryption is only as safe as its keys. This lesson works out how keys are made, kept, used, replaced and destroyed, what hardware protects them, and how an organisation recovers data without trusting any one person too much.
Stage: Think
One key, many years
Think it through
A school keeps one key that unlocks every backup it has ever made. List everything that could go wrong with that key over the years.
- Hint 1: Think from the day it is made to the day it is no longer needed.
Show what happens
It could be guessed if badly made, copied while being handed over, stolen from where it is kept, used for too long, lost when the only person who knows it leaves, or kept after it should have been destroyed. Each stage of a key’s life can fail.
Think it through
Show the answer
Answer: B. Only if it is truly random.
Only if it is truly random. A predictable key can be guessed however long it is.
Stage: Explain
The key lifecycle
A key is not created once and forgotten. It passes through stages, and each stage can fail.
Think it through
Name the stages a key might pass through, from the moment it is created to the moment it is no longer needed.
- Hint 1: What happens to a key between being made and being used?
Show what happens
Generation, distribution, storage, use, rotation, backup or archival, revocation and destruction. The diagram and the school example follow all eight.
Eight stages, from creation to destruction, one at a time. A key is only as strong as its weakest stage.
Stage 1
Step 1 of 8: Generation
Create the key from unpredictable random values, at a suitable length.
Stage 2
Step 2 of 8: Distribution
Get the key to the people or systems that need it, without exposing it on the way.
Stage 3
Step 3 of 8: Storage
Keep the key where only authorised use is possible, ideally in protected hardware and never in plain text beside the data.
Stage 4
Step 4 of 8: Use
Use the key only for its intended purpose, such as signing or encryption, not both.
Stage 5
Step 5 of 8: Rotation
Replace the key with a new one on a planned schedule, or at once if it may have been exposed.
Stage 6
Step 6 of 8: Backup or archival
Keep a protected copy if data encrypted with the key must remain readable later.
Stage 7
Step 7 of 8: Revocation
Withdraw a key that is compromised or no longer trusted, and tell those who rely on it.
Stage 8
Step 8 of 8: Destruction
Securely erase the key, and every copy, when nothing needs it any more.
A fictional school encrypts its nightly backups. Follow the one key that protects them through all eight stages.
- Generation. The backup system creates a random 256-bit key.
- Distribution. The key is passed to the backup server over a protected connection, and to nothing else.
- Storage. It is held in the school’s key management system, not in a file beside the backups.
- Use. It encrypts backups and is used for no other purpose.
- Rotation. Each September a new key takes over for new backups.
- Backup or archival. Last year’s key is kept, protected, for as long as last year’s backups must stay readable.
- Revocation. A technician’s laptop that could reach the key is lost, so the key is withdrawn at once and replaced.
- Destruction. When the oldest backups pass their retention date and are deleted, the key that protected them is securely erased.
A mistake at any one stage, such as a key left in a file at stage 3 or a key that is never replaced at stage 5, undoes the care taken at the others.
Two stages deserve a comment.
Check your understanding
Show the answer
Answer: B. Revocation
Revocation: the key is withdrawn at once and replaced, without waiting for the scheduled rotation.
Key rotation limits damage. If one key protects ten years of data, one theft exposes ten years. Changing keys regularly limits how much any single key protects and for how long a stolen one is useful.
Destruction can be deliberate protection. If data is encrypted and every copy of the key is destroyed, the data is unreadable for good. This is sometimes used to dispose of data securely, and it is also why losing a key by accident is so serious.
Stage: Predict
Entropy and randomness
A key must be unpredictable. An attacker who can predict how a key was chosen does not need to search the keyspace.
Think it through
Show the answer
Answer: C. About 86,400
About 86,400, one for each second in a day: a moment’s work, whatever the key length.
Entropy is the measure of unpredictability. A 256-bit key should contain 256 bits of entropy, meaning each bit is as likely to be 0 as 1 and tells you nothing about the others.
Computers are built to be predictable, which makes true randomness hard for them. Operating systems therefore gather entropy from physical events, such as the precise timing of hardware activity, and many processors and security chips contain a hardware random number generator. Cryptographic software must use a generator designed for the purpose, called a cryptographically secure random number generator, and not the ordinary “random” function of a programming language, whose output can be predicted.
Imagine a program that creates 256-bit keys but, through a mistake, bases each one on nothing more than the time of day in whole seconds. There are 86,400 seconds in a day. An attacker who knows roughly when a key was made has at most 86,400 candidates to try, not 2256. The key is 256 bits long and has about 16 bits of entropy.
This is not only a thought experiment. In 2008, a flaw in one widely used Linux distribution’s cryptographic library was found to have limited the keys it generated to a few tens of thousands of possibilities, and every affected key had to be replaced.
Length is not strength unless the key is random.
Stage: Predict
Centralised and decentralised key management
Who creates the keys, and where are they kept?
- In centralised key management, a dedicated key management system generates, stores, rotates and issues keys for the whole organisation. Applications ask it for a key when they need one.
- In decentralised key management, each system, application or person generates and looks after their own keys.
Think it through
Show the answer
Answer: B. Centralised.
Centralised key management, in which the organisation holds and can recover the keys.
| Aspect | Centralised | Decentralised |
|---|---|---|
| Control and audit | One place to apply policy and to log every use | Varies with each owner |
| Recovery | The organisation can recover data when a person leaves | A lost key may mean lost data |
| Consistency | Rotation and key lengths enforced everywhere | Depends on each owner remembering |
| Main risk | A single, very valuable target and point of failure | Keys scattered, forgotten or poorly stored |
Most organisations of any size centralise, and then protect the central system very carefully.
KMIP
A key management system is useful only if products can talk to it. The Key Management Interoperability Protocol (KMIP) is an open standard, maintained by the standards body OASIS, that defines how a client such as a storage array, a database or a backup system asks a key management server to create, fetch, rotate or destroy a key. With KMIP, one key management server can serve equipment from many vendors.
Stage: Predict
Hardware that protects keys
Think it through
A key is stored in an ordinary file on a server. Who could copy it?
- Hint 1: Who and what can read files on a server?
Show what happens
Anything that can read the file: malware, a dishonest administrator, or anyone who gets a backup of the disk. That is why the stronger approach keeps keys inside hardware they cannot leave.
A key stored in an ordinary file can be copied by anything that can read the file, including malware and a dishonest administrator. The stronger approach is to keep keys inside hardware that will use a key on request and will not hand it over.
Trusted Platform Module
A Trusted Platform Module (TPM) is a small security chip built into a computer’s motherboard, or implemented in its processor firmware. It belongs to that one device.
- It generates keys and stores them so that they cannot be copied off the device.
- It can measure how the computer starts and release a key only if the device has started in the expected, untampered state.
- It resists guessing: after repeated wrong PINs it locks out further attempts.
Its best-known use is with full-disk encryption such as BitLocker, described in Data States and Encryption at Rest.
Check your understanding
Show the answer
Answer: C. One device
One device: it is built into, or implemented in, that computer.
A TPM does not hold every key
It is often said that “the disk encryption key is in the TPM”. That is not quite right, and the difference matters when something goes wrong.
In BitLocker, the data is encrypted with one key. That key is itself encrypted by a second key, and both are stored on the drive, in encrypted form. The TPM’s job is to protect, and release in the right conditions, the secret that unlocks the second key. So the TPM guards the top of a chain; it does not contain every key in it.
| Link in the chain | What it does | Where it is kept |
|---|---|---|
| The TPM’s secret | Unlocks the second key, and only if the computer started up untampered | Inside the TPM chip |
| The second key | Unlocks the data key | On the drive, encrypted |
| The data key | Encrypts and decrypts everything on the drive | On the drive, encrypted |
| The recovery key | A second way to unlock the second key, without the TPM | Wherever the organisation stores it safely, away from the computer |
Check your understanding
Show the answer
Answer: A. With the recovery key.
With the recovery key, which unlocks the same second key without the TPM.
This layering is deliberate. It is why a recovery key can unlock the drive when the TPM cannot, for example after the motherboard is replaced: the recovery key unlocks the same chain by another route.
Stage: Predict
Hardware security modules
Think it through
Where would you keep a certificate authority’s signing key, which could be used to issue certificates for any website?
- Hint 1: How could you use a key without anyone ever seeing it?
Show what happens
Inside a dedicated, tamper-resistant device that creates the key, uses it on request, never lets it out in usable form and erases it if tampered with.
A hardware security module (HSM) is a dedicated, tamper-resistant device, or a cloud service backed by such devices, whose only job is to generate, store and use keys. It serves many systems across a network.
- Keys are created inside the HSM and are designed never to leave it in usable form.
- Applications send it data to sign or decrypt, and receive the result.
- Tampering with the device causes it to erase its keys.
- It is built for high volume and for strict separation of administrator duties.
HSMs protect the most valuable keys an organisation has: a certificate authority’s signing keys, a bank’s payment keys, the master keys of a key management system.
Secure enclaves
A secure enclave is an isolated, protected area inside a processor or a system-on-chip. It runs separately from the main operating system, so that even software with full control of the device cannot read what is inside it.
Phones use a secure enclave to hold fingerprint or face data and the keys that unlock the device. Server processors offer enclaves, often called trusted execution environments, in which sensitive data can be processed while staying protected from the rest of the machine, which helps to protect data in use.
Do not confuse: TPM, HSM and secure enclave
- ServesTPM
- One device. A chip built into a computer that protects that computer’s keys and checks how it started.Releasing a laptop’s disk-encryption key at start-up.
- ServesHSM
- Many systems. A dedicated tamper-resistant appliance or service that holds and uses high-value keys.Holding a certificate authority’s signing key.
- ServesSecure enclave
- Code and data on one processor. An isolated area kept apart from the main operating system.Holding a phone’s fingerprint data and unlock keys.
Stage: Predict
FIPS 140-2 and FIPS 140-3
How does a buyer know that a cryptographic product is properly built? One answer is independent testing against a published standard.
FIPS 140 is the United States government standard for cryptographic modules: the hardware or software component that actually performs cryptography. Laboratories test a module against it, and a joint United States and Canadian programme issues a validation certificate.
- FIPS 140-2 was published in 2001.
- FIPS 140-3 was approved in 2019 and supersedes FIPS 140-2. New modules are validated against FIPS 140-3. Validations under FIPS 140-2 stopped being accepted for new certificates, and in September 2026 the remaining FIPS 140-2 certificates were moved to the programme’s historical list.
Think it through
Show the answer
Answer: C. Not necessarily.
Not necessarily. Using approved algorithms is not the same as having the module tested and validated.
Two cautions when reading a product’s claims:
- A validation belongs to a specific module and version, not to a company or a whole product. “FIPS validated” should be followed by a certificate number that can be looked up.
- “FIPS compliant” or “uses FIPS-approved algorithms” is not the same as validated. Using AES does not mean the implementation was tested.
FIPS 140 is required for United States federal systems and widely used elsewhere as a benchmark. It is not a United Kingdom legal requirement.
Go deeper The four security levels
FIPS 140 defines four levels of increasing strictness.
- Level 1: approved algorithms, correctly implemented. No physical protection required.
- Level 2: adds evidence of tampering, such as seals, and role-based access.
- Level 3: adds resistance to tampering, identity-based authentication, and separation between the interfaces through which keys and other data pass.
- Level 4: adds protection against sophisticated physical attacks and environmental manipulation.
HSMs are commonly validated at Level 3.
Stage: Predict
Key escrow
Strong encryption has an uncomfortable consequence: lose the key and the data is gone. Nobody can help, and that includes the vendor.
Think it through
One administrator set up a school’s encrypted file server and is the only person who knows the recovery key. What could go wrong, and how could it be prevented?
- Hint 1: What happens if that person leaves?
Show what happens
If the administrator leaves or is unavailable, the data may be lost for good. A copy of the key placed in advance with a trusted party, under controlled access, prevents that.
Key escrow means placing a copy of a key with a trusted party, so that the data can be recovered if the key, or the person who holds it, is not available.
A fictional school encrypts its file server. One network administrator set it up and is the only person who knows the recovery key. The administrator leaves at short notice. Months later a hardware fault means the server will not unlock by itself and asks for the recovery key.
Without escrow: nobody has the key. Years of pupils’ work and school records are unreadable, permanently.
With escrow: when the server was encrypted, a copy of the recovery key was stored in the school’s directory service and a printed copy was sealed in the school safe, with access recorded. The headteacher authorises retrieval, the server is unlocked, and a new recovery key is generated because the old one has now been seen.
A spare house key left with a trusted neighbour is a reasonable comparison, with limits. A copied digital key is a perfect copy that can be copied again without trace, and whoever holds it can read the data without ever entering the building. So escrow moves risk; it does not remove it.
- Who holds the copy? An internal team, a directory service, or an outside provider. They must be at least as trustworthy and as well protected as the original holder.
- Who may retrieve it, and how is that recorded? Retrieval should need authorisation and leave an audit record.
- Which keys are escrowed? Keys that encrypt stored data are escrowed, because losing them loses the data. Keys used for signing generally are not: if a second copy of a signing key exists, a signature no longer points at one person, which weakens non-repudiation.
Check your understanding
Show the answer
Answer: B. Signing keys
Signing keys: a second copy would mean the owner is no longer the only person who could have signed.
Stage: Predict
M-of-N control
Escrow raises a question: what stops the one person who holds the escrowed copy from misusing it?
Think it through
What stops the one person who holds an escrowed key from misusing it?
- Hint 1: Could you require more than one person?
Show what happens
Nothing, unless the authority is split. Requiring several people to act together means no one person can use it alone.
M-of-N control splits the authority among N people so that any M of them must act together. The recovery key is divided into shares, and fewer than M shares reveal nothing.
Suppose the school uses 3 of 5. Five trustees each hold a share: the headteacher, the business manager, the IT lead, the chair of governors and a deputy head.
- No one person can recover the data alone, and nor can any two acting together.
- The school survives absences: any two of the five can be ill, on holiday or have left, and recovery is still possible.
Check your understanding
Show the answer
Answer: C. Two
Two: any three of the five can still recover the key.
Choosing the numbers is a balance. A higher M means more people must agree, which is safer and less convenient. A larger gap between M and N tolerates more absences.
M-of-N control is how the most sensitive keys are handled. A certificate authority’s root key is typically activated only when several named officers are present, each with their own share. It is separation of duties applied to a key.
Stage: Explain
Security+ terminology
The CompTIA Security+ SY0-701 exam objectives (objective 1.4) list key escrow under public key infrastructure, and under Tools: Trusted Platform Module (TPM), hardware security module (HSM), key management system and secure enclave.
The objectives do not list KMIP, entropy, FIPS 140 or M-of-N control by name; they explain how the listed tools are used and assured.
Stage: Apply
Apply it in a new situation
These questions use the lesson in situations it has not shown. For the strongest test of your understanding, come back to them on another day and answer each one at the first attempt.
Apply it
Show the answer
Answer: B. Rotate: one theft exposes all.
One theft would expose ten years of backups. Rotate keys regularly so that each protects less, for less time.
Apply it
Show the answer
Answer: B. The certificate number
The validation certificate number, for that module and version, so that it can be looked up.
Apply it
Show the answer
Answer: B. One absence blocks it.
If either trustee is unavailable, the key cannot be recovered at all. A gap between M and N tolerates absences.
Stage: Check
Check your understanding
These questions come from the approved revision bank. They were approved under owner-authorised delegated review: the site owner authorised the review and did not read each question individually.
Key lifecycle, randomness and escrow
Show answer
Answer: Destruction
The cryptographic key lifecycle runs through generation, distribution, storage, use, rotation, backup or archival, revocation and destruction. Securely wiping keys that are no longer needed is destruction. Rotation would replace the keys with new ones for continued use, distribution delivers keys to where they are needed, and generation creates them.
Revision sources: Session 9: Cryptographic Algorithms, Public Key Infrastructure and Cryptographic Solutions (Classroom session record) · Key Management, Cryptographic Hardware and Escrow (Supplementary Study Guide explanation)
Show answer
Answer: Replacing it with a new key on a planned basis or after suspected exposure
Rotation replaces a key with a new one on a planned basis or after suspected exposure, which limits how much data any one key protects and for how long. Copying a key to a trusted party is escrow, and dividing the authority to use it among several people is M-of-N control. Moving a key around does not replace it.
Revision sources: Key Management, Cryptographic Hardware and Escrow (Supplementary Study Guide explanation)
Show answer
Answer: False
False. Keys must be generated from unpredictable random values, and a key generated from too little entropy can be guessed however long it is. If a generator can only ever produce a few thousand different outputs, an attacker tries those few thousand, not the full 256-bit keyspace.
Revision sources: Key Management, Cryptographic Hardware and Escrow (Supplementary Study Guide explanation)
Show answer
Answer: Centralised key management
Centralised key management generates and stores keys in a dedicated system that issues them to the services that need them, while decentralised management leaves each system or user to generate and hold its own. Escrow and M-of-N control concern recovering a key, not where keys are generated and issued.
Revision sources: Session 9: Cryptographic Algorithms, Public Key Infrastructure and Cryptographic Solutions (Classroom session record) · Key Management, Cryptographic Hardware and Escrow (Supplementary Study Guide explanation)
Show answer
Answer: Letting products from different vendors communicate with a key management server
The Key Management Interoperability Protocol (KMIP) is a standard protocol that lets applications and devices from different vendors communicate with a key management server, so that one system can serve many products. Unpredictability of random values is entropy, dividing recovery among people is M-of-N control, and a published list of withdrawn certificates is a revocation list.
Revision sources: Session 9: Cryptographic Algorithms, Public Key Infrastructure and Cryptographic Solutions (Classroom session record) · Key Management, Cryptographic Hardware and Escrow (Supplementary Study Guide explanation)
Show answer
Answer: Key escrow: a copy of the key held by a trusted party
Key escrow places a copy of a key with a trusted party so that data can be recovered if the key or its holder is unavailable, which is the school’s position. Rotation, destruction and generation are other lifecycle stages, and none of them leaves the school with a recoverable copy of a key that only one departed person knew.
Revision sources: Session 9: Cryptographic Algorithms, Public Key Infrastructure and Cryptographic Solutions (Classroom session record) · Key Management, Cryptographic Hardware and Escrow (Supplementary Study Guide explanation)
Show answer
Answer: True
True. An escrowed copy is another copy to protect, so access to it must be controlled and recorded. Escrow trades a recovery risk for a disclosure risk: whoever can reach the escrowed copy can decrypt the data, which is why escrow is often combined with M-of-N control.
Revision sources: Key Management, Cryptographic Hardware and Escrow (Supplementary Study Guide explanation)
Show answer
Answer: No single trustee can recover the data alone, and two absences can be tolerated
M-of-N control splits the authority to recover or use a key among N people so that any M of them must act together. With three of five, one person cannot act alone and the loss of two does not block recovery. The arrangement does not lengthen the key, no trustee holds a full copy, and a lost key cannot be recreated from new random values.
Revision sources: Session 9: Cryptographic Algorithms, Public Key Infrastructure and Cryptographic Solutions (Classroom session record) · Key Management, Cryptographic Hardware and Escrow (Supplementary Study Guide explanation)
TPM, HSM, secure enclave and FIPS 140
Show answer
Answer: A TPM protects keys for the one device it is built into; an HSM serves many systems
A TPM is a security chip built into one device that generates and protects keys for that device, while an HSM is a dedicated, tamper-resistant device or service that generates, stores and uses keys on behalf of many systems. An isolated area inside a processor is a secure enclave, and the standard that cryptographic modules are validated against is FIPS 140.
Revision sources: Session 9: Cryptographic Algorithms, Public Key Infrastructure and Cryptographic Solutions (Classroom session record) · Key Management, Cryptographic Hardware and Escrow (Supplementary Study Guide explanation)
Show answer
Answer: A hardware security module
A hardware security module is a dedicated, tamper-resistant device or service that generates, stores and uses keys on behalf of many systems, such as a certificate authority’s signing keys, without releasing them. A TPM protects keys for the single device it is built into, a phone’s secure enclave serves that phone, and a FIPS 140 certificate records a validation and stores nothing.
Revision sources: Key Management, Cryptographic Hardware and Escrow (Supplementary Study Guide explanation)
Show answer
Answer: The TPM releases its protected key only on the original device in its expected state
A TPM is built into one device, protects keys for that device and can release them only when the device starts in an expected state. The drive’s new host has a different TPM, or none, so the protecting key is not released and the recovery key is needed. A TPM does not delete keys at power-off, re-encrypt drives or hold the files themselves.
Revision sources: Key Management, Cryptographic Hardware and Escrow (Supplementary Study Guide explanation)
Show answer
Answer: False
False. A TPM does not necessarily hold every encryption key, and commonly protects a key that in turn protects others stored on the drive. The key that encrypts the data can sit on the drive in encrypted form, unreadable until the TPM releases the key that protects it.
Revision sources: Key Management, Cryptographic Hardware and Escrow (Supplementary Study Guide explanation)
Show answer
Answer: Keys and sensitive processing, kept isolated from the main operating system
A secure enclave is an isolated, protected area of a processor or system-on-chip that keeps keys and sensitive processing apart from the main operating system, protecting data while it is in use. Encrypting a whole drive in hardware, agreeing connection keys and publishing withdrawn certificates are the jobs of other technologies.
Revision sources: Session 9: Cryptographic Algorithms, Public Key Infrastructure and Cryptographic Solutions (Classroom session record) · Key Management, Cryptographic Hardware and Escrow (Supplementary Study Guide explanation)
Show answer
Answer: FIPS 140-3 supersedes FIPS 140-2 as the standard for validating cryptographic modules
FIPS 140 is the United States government standard for validating cryptographic modules, and FIPS 140-3 supersedes FIPS 140-2. It is a standard that modules are tested against, not a cipher, and the protocol that lets key management products interoperate is KMIP.
Revision sources: Session 9: Cryptographic Algorithms, Public Key Infrastructure and Cryptographic Solutions (Classroom session record) · Key Management, Cryptographic Hardware and Escrow (Supplementary Study Guide explanation)
Show answer
Answer: True
True. A FIPS 140 validation applies to a specific module, not to a whole product or organisation. A claim that a company or a service is “FIPS validated” therefore needs reading closely: the question is which module was validated, and whether that module is the one doing the cryptography.
Revision sources: Key Management, Cryptographic Hardware and Escrow (Supplementary Study Guide explanation)
Study and revision
Practise through the Cryptography & Data Protection topic. Questions cover the lifecycle stages, rotation, entropy, centralised management, KMIP, escrow and M-of-N control, and then the differences between a TPM, an HSM and a secure enclave and what FIPS 140 validation covers; ten flashcards recall the main points. Most are learning-only. The questions that have been verified against CompTIA Security+ SY0-701 objective 1.4 are also in the Security+ practice track; that does not mean the lesson covers the whole of an objective, and the lesson does not establish practical experience.
Stage: Consolidate
What to remember
- Keys have a lifecycle, from generation to destruction, and a failure at any stage undoes the rest.
- Length is not strength unless the key is random.
- Centralised key management gives control and recovery, but becomes a valuable target.
- TPMs protect keys on one device; HSMs protect an organisation’s most valuable keys; secure enclaves isolate keys inside a processor.
- Key escrow keeps a protected copy for recovery; M-of-N control stops any one person using it alone.
Key terms
- Key management
Looking after cryptographic keys for their whole lives: generating, distributing, storing, using, rotating, backing up, revoking and destroying them.
- Entropy
The measure of unpredictability in the random values used to generate keys.
- KMIP (Key Management Interoperability Protocol)
A standard protocol that lets applications and devices from different vendors communicate with a key management server.
- TPM (Trusted Platform Module)
A security chip built into one device that generates and protects keys for that device, and can release them only when the device starts in an expected state.
- HSM (Hardware security module)
A dedicated, tamper-resistant device or service that generates, stores and uses keys on behalf of many systems without releasing them.
- Secure enclave
An isolated, protected area of a processor or system-on-chip that keeps keys and sensitive processing apart from the main operating system.
- Key escrow
Placing a copy of a key with a trusted party, so that data can be recovered if the key or its holder is unavailable.
Common misunderstandings
Only if it was generated from enough entropy. A long key from a predictable source can be guessed.
A TPM protects the keys of the one device it is part of. An HSM is a dedicated device that holds and uses keys for many systems.
Not necessarily. It commonly protects a key that unlocks others, which are stored encrypted on the drive.
FIPS 140-3 supersedes it. A validation also applies to one module, not to an organisation.
Escrow is a recovery arrangement chosen by the key’s owner. It does add a copy that must be protected, which is why access to it is controlled and recorded.
Further reading
- NIST SP 800-57 Part 1: Recommendation for Key Management: the key lifecycle and key states.
- NIST: FIPS 140-3 transition effort: the dates on which FIPS 140-3 replaced FIPS 140-2.
- OASIS: Key Management Interoperability Protocol: the body that maintains KMIP.
- Microsoft Learn: BitLocker FAQ: which keys BitLocker uses, where they are stored and where recovery keys can be kept.
- Microsoft Learn: Trusted Platform Module technology overview: what a TPM does in Windows.
- CompTIA Security+ SY0-701 exam objectives (PDF): objective 1.4.
Lesson sources and acknowledgements
Sources and scope
- Covered in Session 9 (see the Session 9 Learning Log): the cryptographic key lifecycle, centralised and decentralised key management and KMIP; entropy and cryptographic randomness; TPMs, HSMs and secure enclaves; FIPS 140-2 and FIPS 140-3; and key escrow and M-of-N threshold recovery.
- Accuracy qualifications: a TPM does not necessarily contain every volume encryption key; FIPS 140-3 supersedes FIPS 140-2; a FIPS 140 validation applies to a specific module.
- Independently checked: the key lifecycle against NIST SP 800-57 Part 1; the BitLocker key arrangement against Microsoft’s BitLocker documentation; the FIPS 140 dates against NIST’s transition page; and the Security+ wording against the published SY0-701 objectives. These are external references, not course evidence.
- Supplementary explanation: every definition, the eight lifecycle stages and their wording, the school backup-key example, the weak-source example, the comparison table, the account of each kind of hardware, the BitLocker key chain, the FIPS levels and cautions, the school escrow example, the house-key comparison and its limits, the table of the links in the BitLocker key chain, the note on how the lesson is laid out, the 3-of-5 example and the misunderstandings. These were developed afterwards and are my explanations, not the tutor’s words. The diagram is original.
- Guided version (published 1 November 2026, awaiting the site owner’s editorial review): the opening questions about a school’s backup key, every question asked in the lesson with its feedback and hints, the application questions and the closing summary are supplementary explanation written for the Study Guide. The settings are fictional. The questions in the lesson’s steps are not revision-bank items; those in “Check your understanding” are, unchanged.
- No practical work is claimed. The school and its staff are fictional. I have not configured a TPM, an HSM, a key management system or key escrow as part of this lesson.
The standard symmetric block cipher, with 128-, 192- or 256-bit keys.
A record of which account did what and when on a system, such as sign-ins, changes and approvals.
The Windows feature that encrypts whole volumes, normally using the computer’s TPM to release the key at start-up.
A certificate authority withdrawing a certificate before its expiry date, for example because its private key was compromised.
A measure intended to modify risk, for example by restricting access, detecting misuse or supporting recovery.
A secret value that controls what a cipher does. The same algorithm with a different key gives a different result.
The hardware or software component that actually performs cryptography, such as an HSM or a software library. It is what a FIPS 140 validation applies to.
The practice of protecting information by transforming it, so that only the intended parties can read it or trust it.
A random number generator designed so that its output cannot be predicted, used to create keys, salts and other secret values.
A document, signed by a certificate authority, that binds a public key to the identity of its subject. The standard format is X.509.
Transforming data with a cipher and a key so that it cannot be read without the right key. Decryption reverses it.
The measure of unpredictability in the random values used to generate keys.
The United States government standard against which cryptographic modules are tested and validated.
The software built into a device to make its hardware work, such as the program inside a camera, printer, router or switch.
Encryption of everything on a drive, including the operating system.
A dedicated, tamper-resistant device or service that generates, stores and uses keys on behalf of many systems without releasing them.
Placing a copy of a key with a trusted party, so that data can be recovered if the key or its holder is unavailable.
The number of bits in a key. Each added bit doubles the number of possible keys.
Looking after cryptographic keys for their whole lives: generating, distributing, storing, using, rotating, backing up, revoking and destroying them.
A dedicated system that generates, stores, rotates and issues keys for an organisation’s applications and devices.
Replacing a key with a new one, on a planned schedule or at once after suspected exposure.
The number of possible keys for a cipher. A key of n bits has 2 to the power n possible values.
A standard protocol that lets applications and devices from different vendors communicate with a key management server.
Splitting the authority to recover or use a key among N people, so that any M of them must act together.
Malicious software designed to disrupt, damage or gain unauthorised access to devices, networks or data. Ransomware is one type.
The software that runs a computer and lets other programs use it. Windows, macOS, Linux, Android and iOS are operating systems.
The certificate authorities, certificates, software and policies that bind public keys to identities and manage certificates through their lives.
A document that sets organisational rules and expectations, defining what is and is not permitted.
A separate key that unlocks an encrypted drive when the normal method cannot, for example after a hardware change. BitLocker’s is a 48-digit number.
Uncertainty and potential harm to an organisation’s objectives, considering what could happen, how likely it is, the consequences and the organisation’s context.
An isolated, protected area of a processor or system-on-chip that keeps keys and sensitive processing apart from the main operating system.
A security chip built into one device that generates and protects keys for that device, and can release them only when the device starts in an expected state.
Your progress on this lesson
Saved in this browser only. It is not sent anywhere, does not follow you to another device, and is lost if you clear this site’s data.
This browser is not allowing the site to save progress, so it will be forgotten when you leave the page.