Long-Term Archive Service Requirements
What a digital archive must do to keep data, and proof of its integrity, for decades.
Wallace, C., Pordesch, U., and R. Brandner, "Long-Term Archive Service Requirements", RFC 4810, DOI 10.17487/RFC4810, March 2007, <https://www.rfc-editor.org/info/rfc4810>.
@misc{rfc4810,
series = {Request for Comments},
number = 4810,
howpublished = {RFC 4810},
publisher = {RFC Editor},
doi = {10.17487/RFC4810},
url = {https://www.rfc-editor.org/info/rfc4810},
author = {C. Wallace and U. Pordesch and R. Brandner},
title = {{Long-Term Archive Service Requirements}},
pagetotal = 17,
year = 2007,
month = mar,
} RFC text
Contents
Network Working Group C. Wallace
Request for Comments: 4810 Cygnacom Solutions
Category: Informational U. Pordesch
Fraunhofer Gesellschaft
R. Brandner
InterComponentWare AG
March 2007
Long-Term Archive Service Requirements
Status of This Memo
This memo provides information for the Internet community. It does
not specify an Internet standard of any kind. Distribution of this
memo is unlimited.
Copyright Notice
Copyright (C) The IETF Trust (2007).
Abstract
There are many scenarios in which users must be able to prove the
existence of data at a specific point in time and be able to
demonstrate the integrity of data since that time, even when the
duration from time of existence to time of demonstration spans a
large period of time. Additionally, users must be able to verify
signatures on digitally signed data many years after the generation
of the signature. This document describes a class of long-term
archive services to support such scenarios and the technical
requirements for interacting with such services.
Table of Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 3
2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 4
3. General Principles . . . . . . . . . . . . . . . . . . . . . . 5
4. Technical Requirements . . . . . . . . . . . . . . . . . . . . 6
4.1. Enable Submission, Retrieval, and Deletion of Archived
Data Objects . . . . . . . . . . . . . . . . . . . . . . . 6
4.1.1. Functional Requirements . . . . . . . . . . . . . . . 7
4.1.2. Rationale . . . . . . . . . . . . . . . . . . . . . . 7
4.2. Operate in accordance with a long-term archive policy . . 8
4.2.1. Functional Requirements . . . . . . . . . . . . . . . 8
4.2.2. Rationale . . . . . . . . . . . . . . . . . . . . . . 9
4.3. Enable Management of Archived Data Objects . . . . . . . . 9
4.3.1. Functional Requirements . . . . . . . . . . . . . . . 9
4.3.2. Rationale . . . . . . . . . . . . . . . . . . . . . . 9
4.4. Provide Evidence Records that Support Demonstration of
Data Integrity . . . . . . . . . . . . . . . . . . . . . . 10
4.4.1. Functional Requirements . . . . . . . . . . . . . . . 10
4.4.2. Rationale . . . . . . . . . . . . . . . . . . . . . . 10
4.5. Support Data Confidentiality . . . . . . . . . . . . . . . 11
4.5.1. Functional Requirements . . . . . . . . . . . . . . . 11
4.5.2. Rationale . . . . . . . . . . . . . . . . . . . . . . 11
4.6. Provide Means to Transfer Data and Evidence from One
Service to Another . . . . . . . . . . . . . . . . . . . . 11
4.6.1. Functional Requirements . . . . . . . . . . . . . . . 11
4.6.2. Rationale . . . . . . . . . . . . . . . . . . . . . . 11
4.7. Support Operations on Groups of Data Objects . . . . . . . 12
4.7.1. Functional Requirements . . . . . . . . . . . . . . . 12
4.7.2. Rationale . . . . . . . . . . . . . . . . . . . . . . 12
5. Operational Considerations . . . . . . . . . . . . . . . . . . 12
6. Security Considerations . . . . . . . . . . . . . . . . . . . 13
7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 14
8. Informative References . . . . . . . . . . . . . . . . . . . . 14
Appendix A. Application Scenarios . . . . . . . . . . . . . . . . 15
A.1. Archive Service Supporting Long-Term Non-Repudiation . . . 15
A.2. Pure Long-Term Non-Repudiation Service . . . . . . . . . . 15
A.3. Long-Term Archive Service as Part of an Internal
Network . . . . . . . . . . . . . . . . . . . . . . . . . 15
A.4. Long-Term Archive External Service . . . . . . . . . . . . 15
1. Introduction
Digital data durability is undermined by continual progress and
change on a number of fronts. The useful lifetime of data may exceed
the life span of formats and mechanisms used to store the data. The
lifetime of digitally signed data may exceed the validity periods of
public-key certificates used to verify signatures or the
cryptanalysis period of the cryptographic algorithms used to generate
the signatures, i.e., the time after which an algorithm no longer
provides the intended security properties. Technical and operational
means are required to mitigate these issues. A solution must address
issues such as storage media lifetime, disaster planning, advances in
cryptanalysis or computational capabilities, changes in software
technology, and legal issues.
A long-term archive service aids in the preservation of data over
long periods of time through a regimen of technical and procedural
mechanisms designed to support claims regarding a data object. For
example, it might periodically perform activities to preserve data
integrity and the non-repudiability of data existence by a particular
point in time or take actions to ensure the availability of data.
Examples of periodic activities include refreshing time stamps or
transferring data to a new storage medium.
A long-term archive service may be used to provide evidence that
supports validation of the existence of documents or assertions of
agreements that were originally asserted with digital signatures.
Validation may occur at times in the future well beyond the validity
period of the private key originally used to generate the signature,
or even beyond the time when the algorithms available for digital
signatures, message digesting, or data encryption cease to offer
effective protection because of improvements in computing speeds and
methods.
A long-term archive service may be located within an enterprise
network, communicating with local storage mechanisms and other
applications, or a long-term archive service may be implemented as an
external service accessible via the Internet. A long-term archive
service may use functionality, e.g., time stamping, provided by
independent service providers.
A primary goal of a long-term archive service is to support the
credible assertion of a claim that is currently asserted, at points
well into the future. A long-term archive service may support a
range of applications, including: wills, land records, medical data,
criminal case files, personnel files, and contracts. A long-term
archive service may be used by any type of entity, e.g.,
organizations, citizens, notaries. Examples of long-term archive
service usage by submitters include:
- A company stores contracts using a third party service.
- A hospital stores medical data using an internal service.
- An individual wants to generate evidence of data possession at a
particular point in time, e.g., for intellectual property purposes
or endorsement of a contract.
- A law enforcement officer wants to store criminal data such that
integrity of the data can be demonstrated years later.
For each of the above examples, there is a corresponding example
involving retrievers, e.g., a company retrieves a contract in the
case of a dispute or a law enforcement officer prepares information
for a criminal trial.
This document addresses the technical requirements for a long-term
archive service.
2. Terminology
We define the following terms based on their usage in the archiving
community, in order to provide a vocabulary for describing
requirements and the standards around them.
Arbitrator: Principal for whom the validity of archived data
characteristics, e.g., origin, integrity or time of existence,
must be demonstrated.
Archival Period: The period during which an archived data object is
preserved by a long-term archive service.
Archived Data Object: Data unit to be preserved by a long-term
archive service.
Archive Package: Collection of information including archived data
objects and associated Evidence Record.
Cryptographic Maintenance Policy: A set of rules that defines how
to maintain the validity of digitally signed objects should one of
the hash or asymmetric algorithms used to create a digital
signature become weak, or one of the private keys used to create a
digital signature be compromised or become weak.
Evidence: Information that may be used to demonstrate the validity
of an archived data object or related attestations.
Evidence Record: Collection of evidence compiled for one or more
archived data objects. An Evidence Record may include
acknowledgements from a long-term archive service, time stamps and
verification data, such as public-key certificates, revocation
information, trust anchors, policy details and role information.
Long-Term Archive Policy: A set of rules that define operational
characteristics of a long-term archive service.
Long-Term Archive Service (LTA): A service that is responsible for
preserving data for long periods.
Modifier: Principal who modifies attributes associated with an
archived data object and/or Evidence Record held by a long-term
archive service.
Originator: Principal who produces, and possibly digitally signs,
an archived data object. The Originator does not necessarily have
any relationship with a long-term archive service or any awareness
of an Evidence Record associated with the archived data object.
Retriever: Principal who retrieves archived data objects and/or
Evidence Records from a long-term archive service.
Submitter: Principal who submits data objects for archiving.
Time Stamp: An attestation generated by a Time Stamping Authority
(TSA) that a data item existed at a certain time. For example,
[RFC3161] specifies a structure for signed time stamp tokens as
part of a protocol for communicating with a TSA.
Time Stamping Authority (TSA): A trusted service that provides
attestations of existence of data at particular points in time.
For example, [RFC3161] defines protocol elements for interacting
with a TSA.
3. General Principles
A long-term archive service may accept any type of data for
preservation. The data might be in any format, whether textual data,
images, documents, applications, or compound packages of multiple
components. The data may be digitally signed, time stamped,
encrypted, or not subject to any cryptographic processing.
A long-term archive service may preserve archived data objects as
opaque collections of bytes with the primary aim of data integrity.
A long-term archive service is not required to operate upon evidence
related to the content of archived data objects. Content-focused
operations, including data format migration or translation, may be
performed by another service. However, an LTA may incorporate
support for such services.
Different long-term archive services may establish policies and
procedures for archiving data objects over different lengths of time.
For example, an LTA may refuse to preserve archived data objects for
periods longer than 30 years. Similarly, LTAs may establish policies
that limit the types of data that will be accepted for deposit by a
particular LTA.
A long-term archive service provides evidence that may be used to
demonstrate the existence of an archived data object at a given time
and the integrity of the archived data object since that time.
Additionally, the evidence identifies the LTA(s) that have
participated in the preservation of the archived data object. If the
archived data object itself contains digitally signed data,
authentication of the signer is also possible.
A long-term archive service may be an adjunct component of a document
management system. In such cases, the Evidence Record generated and
maintained by the LTA is a property of data that is otherwise managed
by the document management system.
4. Technical Requirements
This section describes the requirements for the protocol for
accessing a long-term archive system and for the data formats
associated with data preservation.
4.1. Enable Submission, Retrieval, and Deletion of Archived Data
Objects
4.1.1. Functional Requirements
A long-term archive service must permit clients to request the
following basic operations:
- submit data objects for archive
- retrieve archived data objects
- delete archived data objects
Following submission, the service must provide an identifier that can
be used to retrieve the archived data and/or associated evidence.
For example, it may be possible to retrieve archive packages by using
a hash value of an archived data object. Possession of this value is
not necessarily an authorization to access the associated archived
data object or evidence record.
It must be possible to authenticate requests and responses, e.g., to
enable LTAs to render an authorization decision. This may be
accomplished by using transport security mechanisms. Requests, in
particular retrieval or deletion requests, may be rejected if the
requestor is not authorized. An authorization policy must be defined
and observed by the long-term archive service. An LTA may disallow
deletion as a matter of policy.
The format for the acknowledgements must allow the identification of
the archiving provider and the participating client.
The LTA must provide an acknowledgement of the deposit that permits
the submitter to confirm the correct data was accepted by the LTA.
This proof need not be provided immediately.
4.1.2. Rationale
Submission, retrieval, query state, and deletion of archived data
objects are necessary basic functions of a long-term archive service.
Deletion may be disallowed due to procedural difficulties in
fulfilling the request. For example, an archived data object may be
stored on write-once media, along with other records that are not
subject to deletion.
Acknowledgements may not be provided immediately due to
implementation of a grace period. A generic query state mechanism
should be provided to address such situations. For example, a
submission response may indicate that a submission has been accepted
and a subsequent query state response may indicate a submission has
completed all necessary preservation steps.
4.2. Operate in accordance with a long-term archive policy
4.2.1. Functional Requirements
A long-term archive service must operate in accordance with a long-
term archive service policy that defines characteristics of the
implementation of the long-term archive service. A long-term archive
service policy contains several components, including:
- Archived data object maintenance policy
- Authorization policy
- Service policy
A long-term archive service policy must include specifications of the
preservation activities performed for archived data objects subject
to the policy. A maintenance policy should define rules for the
following operational aspects: preservation activity triggers,
default archival period, and default handling upon expiration of
archival period.
Maintenance policies should include mechanism-specific details
describing LTA operation. For example, where cryptographic
mechanisms are employed, a cryptographic maintenance policy ought to
be defined.
An authorization policy should define the entities permitted to
exercise services provided by the LTA, including who is permitted to
submit, retrieve, or manage specific archived data objects.
A service policy defines the types of services provided by an LTA,
including acceptable data types, description of requests that may be
accepted, and deletion procedures.
Policies must be unambiguously identified, e.g., by an object
identifier. Alternatively, an LTA may support a protocol that
permits clients to specify policy parameters explicitly instead of by
reference to a policy.
A long-term archive service must be able to provide information
identifying the policies relevant for a given archived data object.
4.2.2. Rationale
Similar to a certificate policies [RFC3647], which are identified
using object identifiers, a long-term archive policy provides a
shorthand means of technically identifying a set of rules that govern
the operation of a long-term archive service.
Over the course of many years, the policies under which an LTA
operates may undergo modification. Thus, an evidence record may
feature multiple indications of policies active at various points
during the life of an archived data object.
4.3. Enable Management of Archived Data Objects
4.3.1. Functional Requirements
A long-term archive service must permit clients to request the
following basic operations:
- specify an archival period for submitted data objects
- extend or shorten the archival period for an archived data object
- specify metadata associated with an archived data object
- specify an archive policy under which the submitted data should be
handled
It should be possible to express an archival period in terms of time,
an event or a combination of time and event.
Submitters should be able to specify metadata that, for example, can
be used to enable retrievers to render the data correctly, to locate
data in an archive or to place data in a particular context.
Examples include, classification codes, type of format, contributors,
title, author, and date. Alternatively, such information may be
included in the content of an archived data object.
If a long-term archive service does not support a requested policy,
it must return an error indication. A service must provide an
indication of the archive policy enforced by the service.
4.3.2. Rationale
Submission, retrieval, and deletion of archived data objects are
necessary basic functions of a long-term archive service.
Specification and management of the archival period is necessary to
avoid unnecessary preservation activities.
4.4. Provide Evidence Records that Support Demonstration of Data
Integrity
4.4.1. Functional Requirements
A long-term archive service must be capable of providing evidence
that can be used to demonstrate the integrity of data for which it is
responsible, from the time it received the data until the expiration
of the archival period of the data.
This may be achieved by providing evidence records that support the
long-term non-repudiation of data existence at a point in time, e.g.,
in the case of legal disputes. The evidence record should contain
sufficient information to enable the validity of an archived data
object's characteristics to be demonstrated to an arbitrator. The
characteristics subject to verification will vary. For example,
authentication of an originator may not be possible in all cases,
e.g., where the object submitted to the archive is not signed or
where the object does not include the necessary information to
authenticate the object's signer.
Evidence records must be structured such that modifications to an
archived data object or its evidence record can be detected,
including modifications made by administrators of an LTA.
4.4.2. Rationale
Supporting non-repudiation of data existence, integrity, and origin
is a primary purpose of a long-term archive service. Evidence may be
generated, or otherwise obtained, by the service providing the
evidence to a retriever. A long-term archive service need not be
capable of providing all evidence necessary to produce a non-
repudiation proof, and in some cases, should not be trusted to
provide all necessary information. For example, trust anchors
[RFC3280] and algorithm security policies should be provided by other
services. An LTA that is trusted to provide trust anchors could
forge an evidence record verified by using those trust anchors.
Demonstration that data has not been altered while in the care of a
long-term archive service is a first step towards supporting non-
repudiation of data. Certification services support cases in which
data must be modified, e.g., translation or format migration. An LTA
may provide certification services.
4.5. Support Data Confidentiality
4.5.1. Functional Requirements
A long-term archive service must provide means to ensure
confidentiality of archived data objects, including confidentiality
between the submitter and the long-term archive service. An LTA must
provide a means for accepting encrypted data such that future
preservation activities apply to the original, unencrypted data.
Encryption, or other methods of providing confidentiality, must not
pose a risk to the associated evidence record.
A long-term archive service should maintain contact information for
the parties responsible for each archived data object so warning
messages can be sent when encryption algorithms require maintenance.
4.5.2. Rationale
Individuals may wish to use the services of a commercial long-term
service without disclosing data to the commercial service. However,
access to the original data may be necessary to perform some
preservation activities.
4.6. Provide Means to Transfer Data and Evidence from One Service to
Another
4.6.1. Functional Requirements
It must be possible to submit data along with previously generated
evidence, i.e., to support transfer of data from one archive to
another. A long-term archive service must support the transfer of
archived data objects, evidence and evidence records from one service
to another. It must be possible for evidence records to span
multiple providers over the course of time, without losing value as
evidence.
4.6.2. Rationale
Before the end of an archived data object's archival period, a long-
term archive service may cease operation. In such cases, it must be
possible for the archived data object (and any associated evidence)
to be transferred to another service that will continue preservation
of the data until the end of the archival period.
Submitters may change service providers before the end of an archived
data object's archival period. In such cases, it must be possible
for the submitter to transfer an archived data object and all
associated evidence from the original LTA to a new LTA.
4.7. Support Operations on Groups of Data Objects
4.7.1. Functional Requirements
An LTA should support submission of groups of data objects.
Submitters should be able to indicate which data objects belong
together, i.e. comprise a group, and retrievers should be able to
retrieve one, some or all members of a group of data objects.
It should be possible to provide evidence for groups of archived data
objects. For example, it should be possible to archive a document
file and a signature file together such that they are covered by the
same evidence record.
Where an LTA operates upon groups of data objects, non-repudiation
proof must still be available for each archived data object
separately.
4.7.2. Rationale
In many cases data objects belong together. Examples include:
- a document file and an associated signature file, which are two
separate objects
- TIF-files representing pages of a document
- a document file and an evidence file (possibly generated by
another LTA)
- a document and its translation to another format or language
In these cases, it is to the best advantage to handle these data
objects as a group.
5. Operational Considerations
A long-term archive service must be able to work efficiently even for
large amounts of archived data objects. In order to limit expenses
and to achieve high performance, it may be desirable to minimize the
use of trusted third parties, e.g., LTA operations should be designed
to limit the number of time stamps required to provide the desired
level of service.
Necessity to access archived data objects should be minimized. It
may only be necessary to access the archived data objects if the
archived data objects are requested by users, or if hash algorithms
used for indexing, or evidence record generation become insecure.
An LTA must be capable of operating in accordance with any applicable
legal regime. For example, an LTA may be required to reject a
deletion request from an authorized requestor if the target of the
request has been subpoenaed by law enforcement authorities.
Some applications may require processing of a chain of archive
policies present in an evidence record, e.g., to ensure that
compatible policies were used throughout the lifetime of the archived
data objects.
6. Security Considerations
Data is the principal asset protected by a long-term archive service.
The principle threat that must be addressed by a long-term archive
service is an undetected loss of data integrity.
In cases where signature verification relies on a PKI, certificate
revocation could retroactively invalidate previously verified
signatures. An LTA may implement measures to support such claims by
an alleged signer, e.g., collection of revocation information after a
grace period during which compromise can be reported or preservation
of subsequent revocation information.
When selecting access control mechanisms associated with data stored
by a LTA, the lifespan of the archived data object should be
considered. For example, the credentials of an entity that submitted
data to an archive may not be available or valid when the data needs
to be retrieved.
During the lifespan of an archived data object, formats may cease to
be supported. Software components to process data, including content
or signatures, may no longer be available. This could be a problem
particularly if non-standard formats are used or proprietary
processing is employed. The submitter should take care to avoid such
problems. For example, the submitter (or other authorized entity)
could periodically retrieve data, convert the data, and re-submit it
in a new format. Additional mechanisms, applications, or tools may
be needed to preserve the value of evidence records associated with
the original archived data object.
A long-term archive system may require correlation of different
identities that represent the same entity at different points in
time. For example, an individual's identity may be represented by
different employers at different points in time.
A long-term archive system must perform maintenance activities on a
schedule that considers factors such as the strength of relevant
cryptographic algorithms, lifespan of relevant certification
authorities, and revocation status of relevant entities, e.g.,
timestamp authorities. Standards for use of cryptographic algorithms
are expected to be established by organization or governmental
bodies, not by individual LTAs.
7. Acknowledgements
Thanks to members of the LTANS mailing list for review of earlier
drafts and many suggestions. In particular, thanks to Larry
Masinter, Denis Pinkas, and Peter Sylvester for review and
suggestions.
8. Informative References
[RFC3161] Adams, C., Cain, P., Pinkas, D., and R. Zuccherato,
"Internet X.509 Public Key Infrastructure Time-Stamp
Protocol (TSP)", RFC 3161, August 2001.
[RFC3280] Housley, R., Polk, W., Ford, W., and D. Solo, "Internet
X.509 Public Key Infrastructure Certificate and
Certificate Revocation List (CRL) Profile", RFC 3280,
April 2002.
[RFC3647] Chokhani, S., Ford, W., Sabett, R., Merrill, C., and S.
Wu, "Internet X.509 Public Key Infrastructure Certificate
Policy and Certification Practices Framework", RFC 3647,
November 2003.
Appendix A. Application Scenarios
Below are several example application scenarios demonstrating one or
more of the basic service features mentioned above.
A.1. Archive Service Supporting Long-Term Non-Repudiation
A long-term archive service may store data objects, such as signed or
unsigned documents, for authenticated users. It may generate time
stamps for these data objects and obtain verification data during the
archival period or until a deletion request is received from an
authorized entity.
A.2. Pure Long-Term Non-Repudiation Service
A long-term archive service may only guarantee non-repudiation of
existence of data by periodically generating time stamps and
obtaining verification data. It stores data objects (e.g., documents
and signatures) locally only for the purpose of non-repudiation and
does not function as a document archive for users. It does not
support retrieval and deletion of data objects.
A.3. Long-Term Archive Service as Part of an Internal Network
A long-term archive service may be part of an enterprise network.
The network provider and archive service may be part of the same
institution. In this case, the service should obtain non-repudiation
evidence from a third party. An internally generated acknowledgement
may be viewed worthless.
A.4. Long-Term Archive External Service
A long-term archive service may be provided over the Internet for
enterprises or consumers. In this case, archiving and providing
evidence (via time stamps or other means) may be adduced by one
organization and its own technical infrastructure, without using
external services.
Authors' Addresses
Carl Wallace
Cygnacom Solutions
Suite 5200
7925 Jones Branch Drive
McLean, VA 22102
Fax: +1(703)848-0960
EMail: cwallace@cygnacom.com
Ulrich Pordesch
Fraunhofer Gesellschaft
Rheinstrasse 75
Darmstadt, Germany D-64295
EMail: ulrich.pordesch@zv.fraunhofer.de
Ralf Brandner
InterComponentWare AG
Otto-Hahn-Strabe 3
Walldorf, Germany 69190
EMail: ralf.brandner@intercomponentware.com
Full Copyright Statement
Copyright (C) The IETF Trust (2007).
This document is subject to the rights, licenses and restrictions
contained in BCP 78, and except as set forth therein, the authors
retain all their rights.
This document and the information contained herein are provided on an
"AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS
OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY, THE IETF TRUST AND
THE INTERNET ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS
OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF
THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED
WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
Intellectual Property
The IETF takes no position regarding the validity or scope of any
Intellectual Property Rights or other rights that might be claimed to
pertain to the implementation or use of the technology described in
this document or the extent to which any license under such rights
might or might not be available; nor does it represent that it has
made any independent effort to identify any such rights. Information
on the procedures with respect to rights in RFC documents can be
found in BCP 78 and BCP 79.
Copies of IPR disclosures made to the IETF Secretariat and any
assurances of licenses to be made available, or the result of an
attempt made to obtain a general license or permission for the use of
such proprietary rights by implementers or users of this
specification can be obtained from the IETF on-line IPR repository at
http://www.ietf.org/ipr.
The IETF invites any interested party to bring to its attention any
copyrights, patents or patent applications, or other proprietary
rights that may cover technology that may be required to implement
this standard. Please address the information to the IETF at
ietf-ipr@ietf.org.
Acknowledgement
Funding for the RFC Editor function is currently provided by the
Internet Society.
ISO Plain Language
On this page
§1Why Long-Term Archives Are Needed
Digital data does not keep its promises on its own. A file's useful life can outlast the format and the storage medium that hold it. A signed document can outlive the certificate used to check the signature, and it can outlive the algorithm that made the signature, which is the point after which that algorithm no longer gives the protection it was designed to give.
A long-term archive service preserves data over long periods through technical and procedural routines, such as refreshing time stamps and moving data to new storage media, so that a claim about the data can still be made in the future. The service can supply evidence that a document existed at a particular time and has not changed since, and it can let signatures be checked years after they were made. The archive may sit inside an organisation's own network or run as an external service reached over the Internet, and it may use time stamping from independent providers.
The RFC lists the kinds of claim the service must support: wills, land records, medical data, criminal case files, personnel files and contracts. Submitters include companies storing contracts with a third party, hospitals storing medical data on an internal service, individuals who want evidence of possessing data at a particular time, and law enforcement officers who need to show integrity years later. In each case there is a matching retriever, such as a company retrieving a contract during a dispute or an officer preparing a criminal trial.
Files and signatures do not last as long as the claims people make about them. This memo describes a service that keeps data safe for decades and keeps the proof that it has not changed.
§2Who Is Involved and What Is Preserved
The RFC borrows the archiving community's vocabulary so that requirements can be stated precisely. A submitter hands data to the archive and a retriever later asks for it back. The originator produced the data and may have signed it, but need not know the archive or its records. An arbitrator is the party who must be satisfied that the data's origin, integrity or time of existence is genuine. A modifier changes attributes attached to an archived object or its evidence record.
The data itself is an archived data object, and the period during which it is kept is the archival period. An archive package holds archived data objects together with the evidence record collected for them. An evidence record may include the archive's acknowledgements, time stamps and verification data such as public-key certificates, revocation information, trust anchors, policy details and role information. A time stamp is an attestation from a Time Stamping Authority that a data item existed at a certain time.
Two sets of rules govern the service. The long-term archive policy defines the operational characteristics of the service. The cryptographic maintenance policy defines how to maintain the validity of digitally signed objects if a hash or signature algorithm becomes weak, or if one of the private keys used to create a signature is compromised.
The memo names the people involved: the one who hands data over, the one who made it, the one who asks for it later, and the one who has to be convinced the data is genuine. It also names the things: the stored data, its evidence record, and the rules the archive works under.
- SubmitterHands data objects over to the archive.
- OriginatorProduced the data, and may have signed it.
- RetrieverLater asks the archive for the data and its evidence.
- ArbitratorMust be convinced the origin, integrity or time of existence is genuine.
- ModifierChanges attributes attached to a stored object or its evidence.
§4.1Basic Operations, Policy and Management
Clients must be able to submit data objects for archiving, retrieve them, and delete them. After a submission the service must return an identifier that can be used to fetch the data or its evidence, for example a hash value of the object, although holding that value is not automatically permission to see the data. Requests and responses must be authenticatable, an authorization policy must be defined and observed, and the archive may refuse deletion as a matter of policy. A deposit must be acknowledged so the submitter can confirm the right data was accepted, and a generic query mechanism is needed so that an acknowledgement can be delayed by a grace period without leaving the submitter in doubt.
The service must operate under a long-term archive policy with three parts: an archived data object maintenance policy, an authorization policy and a service policy. The maintenance policy should cover preservation activity triggers, the default archival period and the default handling when that period expires. Policies must be unambiguously identified, for example by an object identifier, and the service must be able to say which policies apply to a given archived object.
Management of archived objects must also be possible. Clients must be able to set, extend or shorten the archival period, which may be expressed in terms of time, an event, or both. They must be able to attach metadata such as classification codes, type of format, contributors, title, author and date. They must be able to choose the policy the data is handled under, and the service must report the policy it enforces and return an error when it cannot support the one requested.
Appendix A sketches four shapes the service can take: one that stores data for authenticated users and gives evidence supporting non-repudiation; a pure non-repudiation service that only generates time stamps and verification data and does not retrieve or delete anything; a service run inside an organisation's own network, which should take its non-repudiation evidence from a third party because an internally generated acknowledgement may be viewed as worthless; and an external service offered over the Internet.
The archive must let clients put data in, take it out and delete it. It must work under a written policy, let clients say how long to keep each item and attach labels to it, and tell clients which policy it actually applied.
- Submit, retrieve, deleteThe three basic operations, with identifiers and authorization.
- Operate under a policyMaintenance, authorization and service policies.
- Manage archived objectsArchival periods, metadata and choice of policy.
- Provide evidence recordsProof of integrity that cannot be quietly altered.
- Support confidentialityEncrypted deposits that still allow preservation work.
- Transfer between servicesMove data and its evidence to another archive.
- Handle groups of objectsRelated files archived and retrieved together.
§4.4Evidence Records, Confidentiality and Groups
The service must be able to provide evidence that demonstrates the integrity of the data it holds, from the moment it received the data until the archival period expires. This may take the form of evidence records supporting long-term non-repudiation of the data's existence at a point in time, as in a legal dispute. The record should carry enough information for an arbitrator to be shown the relevant characteristics, though those vary: authenticating the originator is not always possible, for example when the submitted object is unsigned or lacks the information needed to check the signer. Evidence records must be structured so that modifications to the data or to the record can be detected, including modifications made by the archive's own administrators.
An archive need not supply every piece of evidence a non-repudiation proof would need, and in some cases should not be trusted to. Trust anchors and algorithm security policies are better provided by other services, since an archive that supplied the trust anchors could forge a record that those anchors would verify. Confidentiality must be supported, including between submitter and archive, and the service must accept encrypted data in a way that later preservation work still applies to the original, unencrypted data. Encryption must not put the evidence record at risk, and contact information for the parties responsible for each object should be kept so warning messages can be sent when encryption algorithms need maintenance.
Data and evidence must be transferable from one service to another, so an archive can close before an archival period ends and a submitter can change provider. Evidence records must be able to span several providers over time without losing their value. The service should also handle groups of objects: submitters should be able to say which objects belong together, retrievers should be able to take one, some or all of them, and a single evidence record should be able to cover a document and its signature file. Non-repudiation proof must still be available for each object separately.
The heart of the service is the evidence record: proof that the stored data has not changed since the day it arrived. The archive keeps that proof so carefully that even its own administrators cannot quietly alter it, and it lets clients keep their data private while still doing the maintenance work.
- Submit data object
- Request time stamp
- Time stamp
- Request evidence record
- Archive package
§5Operational and Legal Considerations
An archive has to work efficiently even when it holds a great many archived objects. To limit expense and keep up performance it should minimise the use of trusted third parties, so that the design limits the number of time stamps needed to provide the desired level of service. Necessity of access to archived objects should also be minimised: an object may only need to be opened when a user requests it, or when the hash algorithms used for indexing or for evidence record generation become insecure.
Legal rules constrain the service as well. An archive must be able to operate under any applicable legal regime. For example, it may have to reject a deletion request from an otherwise authorised requestor when the target of the request has been subpoenaed by law enforcement authorities. Some applications need to process a chain of archive policies present in an evidence record, for example to ensure that compatible policies were used throughout the lifetime of the archived data objects.
Keeping a lot of data for a long time costs money, so the archive should use outside helpers sparingly and only open a stored file when it really must. Legal orders can also stop it from deleting something a client has asked it to remove.
§6Security Considerations
Data is the principal asset that a long-term archive service protects, and the principal threat is an undetected loss of data integrity. Where signature verification relies on a public key infrastructure, certificate revocation can retroactively invalidate signatures that were valid before. An archive may support a claim made by an alleged signer by collecting revocation information after a grace period during which a compromise can be reported, or by preserving subsequent revocation information.
Access control choices must account for the lifespan of the archived object: the credentials of the entity that submitted data may no longer be available or valid when the data is finally retrieved. Formats also stop being supported over time, and the software needed to process content or check signatures may no longer be available. This is a particular problem with non-standard formats or proprietary processing, so the submitter should take care to avoid it, for example by periodically retrieving the data, converting it and re-submitting it in a new format. Additional mechanisms, applications or tools may then be needed to preserve the value of the evidence records attached to the original object.
A long-term archive system may also have to correlate different identities that represent the same entity at different points in time, for example an individual recorded under different employers at different times. Maintenance activities must run on a schedule that weighs the strength of the relevant cryptographic algorithms, the lifespan of the relevant certification authorities, and the revocation status of relevant entities such as time stamping authorities. Standards for the use of cryptographic algorithms are expected to be established by organisations or governmental bodies, not by individual archive services.
The danger is that stored data quietly changes and nobody notices. The memo also warns that old certificates, old keys and old file formats can all expire while the data is still needed, so the archive must keep checking, refreshing and re-recording.