[
https://issues.apache.org/jira/browse/HDFS-6134?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14044797#comment-14044797
]
Owen O'Malley commented on HDFS-6134:
-------------------------------------
Mike, I remember you from when I interviewed you.
You are talking about collisions between IV's, not key space. By using either
32 bytes of randomness (if someone is worried about crypto attacks there is no
excuse not to use AES256), there is *NO* possibility of collision even assuming
an insanely bad practice of using a single key version for a huge number of
files. I obviously understand and applied the birthday paradox to get the
numbers.
Note that we *already* have key rolling and the key is already a random string
of bytes. Adding additional layers of randomness just gives the appearance of
more security. That may be wonderful in the closed source security world, but
it actively harmful in open source. In open source having a clear
implementation that is open for inspection is by far the best protection.
Note that the other issue with not using the keys as intended is that many
Hadoop users launch jobs that read millions of files. We can't afford to have
the client fetch a different key for each of those millions of files.
> Transparent data at rest encryption
> -----------------------------------
>
> Key: HDFS-6134
> URL: https://issues.apache.org/jira/browse/HDFS-6134
> Project: Hadoop HDFS
> Issue Type: New Feature
> Components: security
> Affects Versions: 2.3.0
> Reporter: Alejandro Abdelnur
> Assignee: Alejandro Abdelnur
> Attachments: HDFSDataatRestEncryptionProposal_obsolete.pdf,
> HDFSEncryptionConceptualDesignProposal-2014-06-20.pdf
>
>
> Because of privacy and security regulations, for many industries, sensitive
> data at rest must be in encrypted form. For example: the healthÂcare industry
> (HIPAA regulations), the card payment industry (PCI DSS regulations) or the
> US government (FISMA regulations).
> This JIRA aims to provide a mechanism to encrypt HDFS data at rest that can
> be used transparently by any application accessing HDFS via Hadoop Filesystem
> Java API, Hadoop libhdfs C library, or WebHDFS REST API.
> The resulting implementation should be able to be used in compliance with
> different regulation requirements.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
