FWIW, I had a PR to the SSG that enables all the available profiles for
RHEL derivatives even though they will fail the FIPS and vendor supported
tests.
It was quickly dismissed for the reasons already mentioned. It's something
that regularly comes up a few times a year and is generally met with the
same responses. It seems to be something that users and admins actively
want but that the SSG maintainers are are not likely to budge on.
----------
Chuck Atkins
Staff R&D Engineer, Scientific Computing
Kitware, Inc.
(518) 881-1183
On Fri, Mar 6, 2020 at 3:04 PM Trey Henefield <trey.henefield(a)ultra-ats.com>
wrote:
Hi Gabe,
A majority of the RHEL STIG rules can directly map to CentOS, as it is
largely based on RedHat releases.
CentOS does release security patches that align with those released for
RedHat.
I can understand the FIPS and Common Criteria concerns. Let those checks
fail as they may. I was previously an evaluator and consultant for Common
Criteria in a prior life, so I do appreciate the hard effort and extreme
costs involved with that. But moving into a system accredditation
standpoint, it’s a simple IA control in RMF. One of many others which it
can address.
Trust me, I was severely disappointed about how little the large effort of
Common Criteria goes unnoticed in a system accredditation process. There’s
not even a requirement to ensure the system is in its evaluated
configuration. Rather, they are more focused on ensuring it meets the STIG
requirements. But in light of DoD’s stance on the acceptance of Open Source
software, the need for Common Criteria certification seems to be left
behind. It is a disappointing reality, but nonetheless a reality.
For example, we utilize OpenSSH on Windows. We compile it ourselves
utilizing the OpenSSL FIPS module to gain FIPS compliance. While
technically since it offers an abundance of IA capabilities, one would
think that it should also be Common Criteria certified. But that is not the
case.
I understand your reasonings. But the fact of the matter is that CentOS is
approved for use and RHEL STIGs are used as guidance in configuring it to
be compliant.
With that said, there’s no reason why a CentOS profile could not be
maintained within the ComplianceAsCode content.
Best regards,
*Trey Henefield, CISSP*
Cyber Security Manager
Ultra Intelligence & Communications
4101 Smith School Road
Building IV, Suite 100
Austin, TX 78744 USA
T: +1 512 327 6795 ext. 647
M: +1 512 541 6450
ultra.group <
http://www.ultra.group>
*From:* Gabe Alford <redhatrises(a)gmail.com>
*Sent:* Friday, March 6, 2020 1:36 PM
*To:* SCAP Security Guide <scap-security-guide(a)lists.fedorahosted.org>
*Cc:* Trey Henefield <trey.henefield(a)ultra-ats.com>; RADUSINOVIC, IVAN J
GS-12 USAF ACC 805 CTS/DOO <ivan.radusinovic(a)us.af.mil>;
open-scap-list(a)redhat.com; Sherwin Farhang <Sherwin.Farhang(a)newwave.io>;
Matěj Týč <matyc(a)redhat.com>
*Subject:* Re: [Open-scap] SCAP's for CentOS 7 & 8
CentOS STIG profile removal was NOT political in any way. There are
multiple reasons:
- No vendor supports CentOS.
- There is no security support for CVEs.
- There is NO DISA STIG for CentOS.
- RHEL-vendor supported and certified profiles do not map to CentOS... The
CentOS project clearly states this.
- CentOS is not common criteria certified.
- CentOS components are not FIPS validated and using it violates Federal
Law.
- CentOS is not in alignment with STIG requirements because it doesn't
meet them due to not meeting applicable laws and vendor support.
- There are valid cybersecurity reasons for not using CentOS.
Now, if CentOS started to have a security team and handled it's own CVE
content, FIPS-validation, and addressed issues, would be a completely
different story.
This is the same for the OSPP, CUI, etc. profiles as well. In fact, any
profile that uses NIST 800-171 or NIST 800-53, CentOS fails to comply with
and encouraging
the community to support faulty security profiles is bad security practice
in and of itself.
On Fri, Mar 6, 2020 at 10:04 AM Watson Sato <wsato(a)redhat.com> wrote:
On Fri, Mar 6, 2020 at 5:05 PM Trey Henefield <
trey.henefield(a)ultra-ats.com> wrote:
Copy that sir.
Thanks for responding.
I would love to contribute my changes. The problem is by the time we get
the latest release all fixed up and in alignment with STIG requirements, a
new release comes out and completely changes the way content is generated.
For example, the last release changed a bunch of compliance code to now use
templates and jinja. So now its not as simple to merge my existing content
as I now need to learn how things are now done differently, fix issues in
the new process and still get my code to work and be compliant. It is very
tiresome doing this continuously every single release. We can never get
things up to speed and commit changes because by the time we turn around,
everything is yet again drastically changed.
Hello Trey, thanks for sharing your pain.
I'd like to better understand the aspects of your changes, are they only
related to content, like rules and remediation? Or do you also tweak the
build system?
I am all for innovation and finding better ways to accomplish tasks. But
these types of changes need to be less frequently to simply allow the
compliance code to mature.
In most development environments, including our own, there is the concept
of major releases and minor releases. Where major releases incorporate
“major” changes into the software, such as changing how things are done in
the software. Minor releases provide for bug fixes to further improve the
reliability of the the software.
I consider changing how all the code is compiled, adding new processes and
deprecating old processes, to be a major change as it has a cascading
affect for much of the content. Whereas, minor changes would focus on
simply improving the content utilizing the existing processes already in
place.
There is an idea to separate the build system from the benchmark
directory:
https://github.com/ComplianceAsCode/content/issues/5125
Does this idea resonate with you?
Separating the content from the benchmark would minimize issues when
compiling processes change, but it could bring another set of problems,
like versioning, or upgrade paths.
I had brought this up before but it just seems to be ignored.
I really would like to see a better effort from the maintainers of this
content to improve on this issue so that those of us utilizing the content
and improving the content can have ample opportunity to upload these
improvements so that the rest of the community can benefit from these
changes alike.
Best regards,
*Trey Henefield, CISSP*
Cyber Security Manager
Ultra Intelligence & Communications
4101 Smith School Road
Building IV, Suite 100
Austin, TX 78744 USA
T: +1 512 327 6795 ext. 647
M: +1 512 541 6450
ultra.group <
http://www.ultra.group>
*From:* Matěj Týč <matyc(a)redhat.com>
*Sent:* Friday, March 6, 2020 4:13 AM
*To:* Trey Henefield <trey.henefield(a)ultra-ats.com>; RADUSINOVIC, IVAN J
GS-12 USAF ACC 805 CTS/DOO <ivan.radusinovic(a)us.af.mil>; Sherwin Farhang <
Sherwin.Farhang(a)newwave.io>; open-scap-list(a)redhat.com
*Subject:* Re: [Open-scap] SCAP's for CentOS 7 & 8
Hello Trey and others,
first of all, the openscap list is mainly about the scanner, while your
points are about the content, for which I recommend the
scap-security-guide(a)lists.fedorahosted.org mailing list.
Regarding the main topic, I vaguely recall discussions about whether to
have CentOS profiles that would be inevitably failing (due to lack of
certification of scanned systems), or whether to remove profiles that
require certifications to pass altogether. The second approach has won,
possibly driven by fears that the project would keep getting reports that a
rule is failing, so maintainers should do something about it and fix it.
The scanner can't say "I am sorry, your system is not certified, that's
why the rule fails" - anyone using the scanner would see just another rule
failing, and one has to read the guide/report to correctly understand
what's the problem.
On 05. 03. 20 18:16, Trey Henefield wrote:
I don’t have access to it at the moment as I am out of the office this
week and next, but I will provide this info when I return.
I was just as surprised as anyone else, but a customer informed me of
this. Because it is open source software, they were able to do a complete
source code review and approved it based on that, since the DoD opensource
policy permits this now.
We still prefer to use RedHat and bring them allot of business, but it
feels that this open source community is more business driven than
community driven, resulting in these types of decisions. But that should
not be the case for an open source community project.
If any of you don't agree with decisions that were taken, let your take on
this be heard in the appropriate mailing list. Community is inseparable
from participation, maintainers will take shortcuts if there is no visible
pushback against them.
While I am on my soap box, I will just reiterate again, I am not a fan of
the drastic changes that are added every two months to a new release. Major
changes like this and overhauling how compliance code gets built every two
months is just bad. These major changes should be released in a new major
release less frequently (6 months, or even once a year). Rather, the
incrimental releases pushed out every two months should be more based on
bring compliance code up to speed to provide more complete and accurate
coverage of the requirements. Everytime I merge in the latest release with
our code, we spend so much time going back and fixing all the things it
breaks.
The Security Compliance field is gaining momentum, and it will continue to
do so in the foreseeable future. Therefore, the project needs to
accommodate ever-increasing requirements - such as introduction of new
profiles, new products, and even new paradigms. The only way to do so is to
refactor its parts and make them perform better. We are so far very
successful in this effort, and what you describe is the inevitable downside
of it.
So Trey, do you think that your code could be upstreamed? That way, it
would become possible for us to keep it in a good shape at a low cost.
Best regards,
*Trey Henefield, CISSP*
Cyber Security Manager
Ultra Intelligence & Communications
4101 Smith School Road
Building IV, Suite 100
Austin, TX 78744 USA
T: +1 512 327 6795 ext. 647
M: +1 512 541 6450
ultra.group <
http://www.ultra.group>
*From:* RADUSINOVIC, IVAN J GS-12 USAF ACC 805 CTS/DOO
<ivan.radusinovic(a)us.af.mil> <ivan.radusinovic(a)us.af.mil>
*Sent:* Thursday, March 5, 2020 10:54 AM
*To:* Trey Henefield <trey.henefield(a)ultra-ats.com>
<trey.henefield(a)ultra-ats.com>; Sherwin Farhang
<Sherwin.Farhang(a)newwave.io> <Sherwin.Farhang(a)newwave.io>;
open-scap-list(a)redhat.com
*Subject:* RE: SCAP's for CentOS 7 & 8
Hey, while we’re on the topic, does anyone have the link depicting it
being on the APL? Which APL are we talking about?
Thanks!
S/F
Ivan Radusinovic
“R+10”; “Sgt Rad”
ISSE || 805th CTS / DOO
mob. 916-934-6592 || off. 702-652-8499 ||
NIPR ivan.radusinovic(a)us.af.mil
SIPR ivan.j.radusinovic.civ(a)mail.smil.mil
"Comm…. works 60% of the time.. all the time!"
- GySgt McCown
*From:* open-scap-list-bounces(a)redhat.com <
open-scap-list-bounces(a)redhat.com> *On Behalf Of *Trey Henefield
*Sent:* Thursday, March 5, 2020 8:47 AM
*To:* Sherwin Farhang <Sherwin.Farhang(a)newwave.io>;
open-scap-list(a)redhat.com
*Subject:* [Non-DoD Source] Re: [Open-scap] SCAP's for CentOS 7 & 8
This is disappointing as it feels like a political move. There is no
functional reason why it could not be supported. It is approved within DoD
as it is on the approved software list.
Rather than removing its support, let the authorities that be make the
decision as to if it is acceptable for use.
Best regards,
*Trey Henefield, CISSP*
Cyber Security Manager
Ultra Intelligence & Communications
4101 Smith School Road
Building IV, Suite 100
Austin, TX 78744 USA
T: +1 512 327 6795 ext. 647
M: +1 512 541 6450
ultra.group <
http://www.ultra.group>
*From:* open-scap-list-bounces(a)redhat.com <
open-scap-list-bounces(a)redhat.com> *On Behalf Of *Sherwin Farhang
*Sent:* Wednesday, March 4, 2020 3:47 PM
*To:* open-scap-list(a)redhat.com
*Subject:* [Open-scap] SCAP's for CentOS 7 & 8
Hi OpenSCAP team,
I tried following a guide to deploy DISA STIG's from RedHat to CentOS I
was dissapointed this functionality is intentionally not supported
anymore. Can't a message be shot across saying "deploying this does not
make you compliant to DISA STIG because this is CentOS and does not carry
the proper certifications"? I just found this tool and tested it out to
deploy pci-dss as a test to a vm and it was really convenient that it
worked so well. It would be nice if that functionality still worked. Also
when I run CentOS 8 and download the workbench no CentOS profiles are
downloaded but they do download for CentOS - 7 Likely a bug unless your
team is actually just removing all functionality for CentOS 8 moving
forward which is disappointing as well. Whenever I see a project remove
functionality a fork is made and the community splits. I hope to hear back.
Thank you,
-Sherwin
*Sherwin Farhang* | CISSP | GCIH | CCSK
ISSO
6518 Meadowridge Rd. | Ste. 100 | Elkridge, MD 21075
P 443.701.5266| F 1.866.824.8914 | newwave.io <
https://www.newwave.io/>
[image: cidimage011.png(a)01D5E32F.9731D090]
<
https://www.facebook.com/NewWave.Telecom.and.Technologies/>
[image: cidimage012.png(a)01D5E32F.9731D090]
<
https://www.instagram.com/newwave_tech/>
[image: cidimage013.png(a)01D5E32F.9731D090]
<
https://twitter.com/NewWave_Tech>
[image: cidimage014.png(a)01D5E32F.9731D090]
<
https://www.linkedin.com/company/newwave_technologies>
[image: cidimage015.png(a)01D5E32F.9731D090] <
https://newwave.io/>
MBE Certified; GSA Schedule 70
CMMI® Maturity Level 4 Rated SVC | Level 3 Rated DEV
ISO 9001:2015 Certified
*Microsoft* *Partner:* Gold Cloud Platform: Gold Datacenter; Silver
Application Development
Confidentiality Notice: This message and all attachments are for the sole
use of the intended recipient(s) and may contain NewWave Telecom &
Technologies confidential and privileged information or otherwise be
protected by law. This message is intended for use by the individual or
entity to which it is addressed and any unauthorized review, use,
disclosure or distribution is prohibited . If you are not the intended
recipient, please contact the sender by reply e-mail and destroy all copies
of the original message.
*Disclaimer*
The information contained in this communication from *trey.henefield(a)ultra-ats.com
<trey.henefield(a)ultra-ats.com> *sent at 2020-03-05 11:47:08 is private
and may be legally privileged or export controlled. It is intended solely
for use by *open-scap-list(a)redhat.com <open-scap-list(a)redhat.com> *and
others authorized to receive it. If you are not *open-scap-list(a)redhat.com
<open-scap-list(a)redhat.com> *you are hereby notified that any disclosure,
copying, distribution or taking action in reliance of the contents of this
information is strictly prohibited and may be unlawful.
_______________________________________________
Open-scap-list mailing list
Open-scap-list(a)redhat.com
https://www.redhat.com/mailman/listinfo/open-scap-list
_______________________________________________
Open-scap-list mailing list
Open-scap-list(a)redhat.com
https://www.redhat.com/mailman/listinfo/open-scap-list
--
Watson Sato
Security Technologies | Red Hat, Inc
_______________________________________________
scap-security-guide mailing list --
scap-security-guide(a)lists.fedorahosted.org
To unsubscribe send an email to
scap-security-guide-leave(a)lists.fedorahosted.org
Fedora Code of Conduct:
https://docs.fedoraproject.org/en-US/project/code-of-conduct/
List Guidelines:
https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives:
https://lists.fedorahosted.org/archives/list/scap-security-guide@lists.fe...
_______________________________________________
Open-scap-list mailing list
Open-scap-list(a)redhat.com
https://www.redhat.com/mailman/listinfo/open-scap-list