[cabfpub] Send us you list of current problems with the Network Security Guidelines

Peter Bowen pzb at amzn.com
Mon Jun 12 04:01:24 UTC 2017


Below are a few things I’ve seen.  I’m happy to put my name to them, so I’m posting to the public list:

1) Offline CAs are treated the same as online CAs.  For example, a system for CA that is based on HSMs stored in safes theoretically needs individual user logins with multi-factor access control.  For systems where the HSMs has multi person access control, this is highly redundant.

2) Root CAs are not required to be air gapped at all times.

3) The scope is far larger than probably intended — it could be viewed as being as far reaching as including CDNs used to distribute CRLs and OCSP responses which have no ability to generate or modify the responses and systems the relay emails to domain contacts which are outside of the CA system

4) The segmentation requirements are confusing (and possibly contradictory): networks or zones based on their functional, logical, and physical (including location) relationship

5) It assumes passwords are the core authentication credential and does not align with current NIST guidance.  Authentication requirements could probably be put in terms of NIST SP 800-63 AAL.

6) It fails to define “multi-factor authentication”

7) It fails to define “remote” (used as part of “remote administration or access”); Is remote anything other then using a keyboard and monitor physically attached to the system motherboard?

8) Certificate Management System and Security Support System definitions are both very broad.  At least one interpretation prevents usage of any system accessible to persons who are not in Trusted Roles, even if such usage is not critical to system security. For example, the CA might have a corporate policy to send logs to a central log server in addition to CA specific log servers.  It is not clear this is allowed.

9) It has no concept of compensating controls; for example, a CA might want to implement channel authentication as an alternative to physical network segmentation (for example using TLS over VLANs rather than physically segmenting LANs).

The list goes on, but this should be a good start.

Thanks,
Peter

> On Jun 9, 2017, at 2:29 PM, Kirk Hall via Public <public at cabforum.org> wrote:
> 
> Yes, we have also noticed “90 days” versus quarterly – Most quarters have more than 90 days.
>  
> Thanks.
>  
> From: Dean Coclin [mailto:Dean_Coclin at symantec.com <mailto:Dean_Coclin at symantec.com>] 
> Sent: Friday, June 9, 2017 2:09 PM
> To: CA/Browser Forum Public Discussion List <public at cabforum.org <mailto:public at cabforum.org>>
> Cc: Kirk Hall <Kirk.Hall at entrustdatacard.com <mailto:Kirk.Hall at entrustdatacard.com>>
> Subject: [EXTERNAL]Re: [cabfpub] Send us you list of current problems with the Network Security Guidelines
>  
> One specific complaint from the auditors I believe was the specific time requirements in the document. For example, if it said you have to change the password at 90 days, and you did it on day 91, it would be an audit failure. I think Don has better examples but that's one I recall. 
> 
> Sent from my iPhone
> 
> On Jun 9, 2017, at 4:35 PM, Kirk Hall via Public <public at cabforum.org <mailto:public at cabforum.org>> wrote:
> 
> Bruce and I want to collect a preliminary list of current problems with the Network Security Guidelines (technically, the Network and Certificate System Security Requirements), so we can have a good discussion of possible new directions at the upcoming F2F.
>  
> To that end – please send Bruce and me a list of the specific requirements (and/or definitions) in the NetSec requirements that you think are most problematic and which should be changed or dropped.  If possible, give us the following data for each problematic issue:
>  
> 1.       Section or definition of the NetSec Requirements that creates the problem
> 2.       What is the problem?
> 3.       What is a possible solution (drop, amend, supplement), with suggested language.
>  
> Bruce and I will combine all suggestions received and report anonymously to the whole group for a discussion in Berlin.  That may give the new Working Group some useful guidance for its ongoing work after that.
>  
> Thanks.
> <Network_Security_Controls_V1.pdf>
> _______________________________________________
> Public mailing list
> Public at cabforum.org <mailto:Public at cabforum.org>
> https://clicktime.symantec.com/a/1/f8a6ATZl_MnLwe29_m42V-mYnSdtACxRZX0POAv19Vo=?d=Yz3SpbucMfHwCGRoOuzLmlu9Wpr_caTWy3ILlB_IoLHc8KA0wZ9ZIIE0tt6_40GuyUeYNqwIwidNiKaKMu-5OhJUpI-0YmfQlXF6WVerqU-ErugytgRRUvyO4rzY8NCkhG397tCFH2roGFp5G4M7Xr7HurgCIsLKvk_CMVy_W33a8G7xs-zP44TZmNNnklelkmp9rYeMxGizl_l43PaLWEPq3okEBqK1ZLhxicxRW5Q-DJpP7uamThvBEDxql9A4GfwEQidWB4-Z4LlFYTHzdzZ1KHdONbJspvsEhltqHQiSSKHqcjPVfopD6S_b1Il_kJ7UeffHCM2fbGuZM3-ATtW4erUTkCsgco80th_9K1GTEs0Ligy4OOIZyCqOKL88l3ZOGCvzMBdsibgW7vTwBA2LhcZdTn_Nqiq7Lfh2iJXO2hDVk8yYzplKWu7J_J7XdsTZJuufN_61qCXKBjZC5VS3fw%3D%3D&u=https%3A%2F%2Fcabforum.org%2Fmailman%2Flistinfo%2Fpublic <https://clicktime.symantec.com/a/1/f8a6ATZl_MnLwe29_m42V-mYnSdtACxRZX0POAv19Vo=?d=Yz3SpbucMfHwCGRoOuzLmlu9Wpr_caTWy3ILlB_IoLHc8KA0wZ9ZIIE0tt6_40GuyUeYNqwIwidNiKaKMu-5OhJUpI-0YmfQlXF6WVerqU-ErugytgRRUvyO4rzY8NCkhG397tCFH2roGFp5G4M7Xr7HurgCIsLKvk_CMVy_W33a8G7xs-zP44TZmNNnklelkmp9rYeMxGizl_l43PaLWEPq3okEBqK1ZLhxicxRW5Q-DJpP7uamThvBEDxql9A4GfwEQidWB4-Z4LlFYTHzdzZ1KHdONbJspvsEhltqHQiSSKHqcjPVfopD6S_b1Il_kJ7UeffHCM2fbGuZM3-ATtW4erUTkCsgco80th_9K1GTEs0Ligy4OOIZyCqOKL88l3ZOGCvzMBdsibgW7vTwBA2LhcZdTn_Nqiq7Lfh2iJXO2hDVk8yYzplKWu7J_J7XdsTZJuufN_61qCXKBjZC5VS3fw%3D%3D&u=https%3A%2F%2Fcabforum.org%2Fmailman%2Flistinfo%2Fpublic>_______________________________________________
> Public mailing list
> Public at cabforum.org <mailto:Public at cabforum.org>
> https://cabforum.org/mailman/listinfo/public <https://cabforum.org/mailman/listinfo/public>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.cabforum.org/pipermail/public/attachments/20170611/23349ca6/attachment-0003.html>


More information about the Public mailing list