Thread: Update encryption options doc for SCRAM-SHA-256
The following documentation comment has been logged on the website: Page: https://www.postgresql.org/docs/10/static/encryption-options.html Description: Section "18.8. Encryption Options" only mentions MD5 as the password storage encryption mechanism, although PostgreSQL 10 introduced the superior SHA256 - somebody looking at the docs would get a bad idea of PostgreSQL's capabilities...
On 2/2/18 18:42, PG Doc comments form wrote: > The following documentation comment has been logged on the website: > > Page: https://www.postgresql.org/docs/10/static/encryption-options.html > Description: > > Section "18.8. Encryption Options" only mentions MD5 as the password storage > encryption mechanism, although PostgreSQL 10 introduced the superior SHA256 > - somebody looking at the docs would get a bad idea of PostgreSQL's > capabilities... I propose the attached patch. I have combined the password storage and password transmission items, because I don't want to go into the details of how SCRAM works on the wire. -- Peter Eisentraut http://www.2ndQuadrant.com/ PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services
Attachment
Thanks for your attention to this.
I'm definitely not a cryptography expert, but it seems to me that the actual mechanisms (MD5, SHA-256) are more important than the protocols used to negotiate them (SASL, SCRAM). When some security expert unfamiliar with PostgreSQL goes over itss documentation to determine whether it's secure, I think it's important to make sure that the word SHA-256 is actually there.
On Sat, Feb 3, 2018 at 8:30 AM, Peter Eisentraut <peter.eisentraut@2ndquadrant.com> wrote:
On 2/2/18 18:42, PG Doc comments form wrote:
> The following documentation comment has been logged on the website:
>
> Page: https://www.postgresql.org/docs/10/static/encryption- options.html
> Description:
>
> Section "18.8. Encryption Options" only mentions MD5 as the password storage
> encryption mechanism, although PostgreSQL 10 introduced the superior SHA256
> - somebody looking at the docs would get a bad idea of PostgreSQL's
> capabilities...
I propose the attached patch. I have combined the password storage and
password transmission items, because I don't want to go into the details
of how SCRAM works on the wire.
--
Peter Eisentraut http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services