Handling of RSASHA256 and RSASHA512 in BIND 9.6.0 and BIND 9.6.0-P1
Mark Andrews
With upcoming deployment of RSASHA256 to sign the root zone, ISC
would like to remind BIND 9.6.0 and BIND 9.6.0-P1 users that use
DLV, but have not yet upgraded, that they will need to upgrade to
a more recent version of BIND 9.6.x as BIND 9.6.0 and BIND 9.6.0-P1
will not correctly handle RSASHA256 and RSASHA512 signed zones in
DLV.
2579. [bug] DNSSEC lookaside validation failed to handle unknown
algorithms. [RT #19479]
This defect was addressed in BIND 9.6.1.
ISC has arranged for two test zones to be made available which are
signed using the new algorithms which are listed in dlv.isc.org.
You can test whether you can successfully resolve these zones using the
following queries.
dig rsasha256.island.dlvtest.dns-oarc.net soa
dig rsasha512.island.dlvtest.dns-oarc.net soa
--
Mark Andrews, ISC
1 Seymour St., Dundas Valley, NSW 2117, Australia
PHONE: +61 2 9871 4742 INTERNET: ma...@isc.org
Doug Barton
news that BIND 9.6.2 is going to have actual support for
RSASHA{256|512}. My cursory reading of the 9.6.2b1 code does not seem
to indicate that it does, although I would be happy to be proven wrong.
I personally don't think it's reasonable to expect everyone who wants
to validate with BIND to upgrade to 9.7.x for a variety of reasons
that I'd be happy to elucidate if they are not obvious.
Doug
--
Improve the effectiveness of your Internet presence with
a domain name makeover! http://SupersetSolutions.com/
Chris Thompson
>While this reminder is timely and helpful, more welcome would be the
>news that BIND 9.6.2 is going to have actual support for
>RSASHA{256|512}. My cursory reading of the 9.6.2b1 code does not seem
>to indicate that it does, although I would be happy to be proven wrong.
>
>I personally don't think it's reasonable to expect everyone who wants
>to validate with BIND to upgrade to 9.7.x for a variety of reasons
>that I'd be happy to elucidate if they are not obvious.
Quoting from https://lists.isc.org/pipermail/bind-users/2009-October/077853.html
(me)
> Will you be adding RSASHA256 support in the 9.5.x and 9.6.x series? It
> might be a bit optimistic to expect everyone to move to 9.7.x by 2010-07-01,
> if that's when the root zone is going to be *really* signed (with RSASHA256,
> according to current reports).
(Evan Hunt)
> Not 9.5.x, as it lacks NSEC3 support.
>
> Adding SHA-2 to 9.6.x would violate our policy of making major
> functional changes only in major releases, so I don't expect we'll
> do that. Given the odd circumstances you mentioned, I won't say for
> certain that we won't--but I doubt it.
>
> 9.7.0 is going to be final in a little over a month, which is fortunate
> timing.
(But it's not too obvious to me that adding support for a new signing
algorithm should necessarily be considered a "major functional change".)
--
Chris Thompson
Email: ce...@cam.ac.uk
Stephane Bortzmeyer
Doug Barton <do...@dougbarton.us> wrote
> While this reminder is timely and helpful, more welcome would be the
> news that BIND 9.6.2 is going to have actual support for
> RSASHA{256|512}.
No, it won't. Migrating to >= 9.6.1 is necessary to avoid breakage,
not to validate the root.
Doug Barton
> (Evan Hunt)
>> Adding SHA-2 to 9.6.x would violate our policy of making major
>> functional changes only in major releases, so I don't expect we'll
>> do that. Given the odd circumstances you mentioned, I won't say for
>> certain that we won't--but I doubt it.
>>
>> 9.7.0 is going to be final in a little over a month, which is fortunate
>> timing.
>
> (But it's not too obvious to me that adding support for a new signing
> algorithm should necessarily be considered a "major functional change".)
Yes, I remembered Evan's statement from a while back, and didn't
respond at the time because I wanted to think about it some more.
Having thought about it, I agree with you that in my mind it's not a
"major functional change," and I strongly believe that adding support
for it in 9.6 is the right thing to do.
To expand on that a little more (and to slightly agree with Stephane)
it's already been necessary for anyone who wants to _validate_ to have
migrated to 9.6 for some time now. 9.6 has proven to be a good
release, and everyone that I've recommended upgrading to it has been
thoroughly satisfied. Therefore (within the "validator" demographic)
we've got a pretty good installed base for whom a minor version
upgrade would not be a problem, and will likely happen when 9.6.2 is
released in any case. Expecting that installed base to upgrade to an
unproven .0 release with a lot of new features (read, untried code
paths) is not realistic. And it should go without saying that this is
with all due respect to the fine people who actually write BIND code.
I know they work hard to get it right, but I also know we're _all_ human.
OTOH for those that want to _sign_ their zones I'm have been telling
people for a while now that they need to start working with 9.7. I
even created a FreeBSD port for the RC version (which I have not done
for previous RCs) to help accelerate that process.
BIND 9.6.2 is in the "b1" phase atm, which means that there is plenty
of time to get SHA2 in there and get the release out before a signed
root goes live. I encourage the folks at ISC to do so, and if you
agree I encourage you to make your voice heard.
Evan Hunt
> of time to get SHA2 in there and get the release out before a signed
> root goes live. I encourage the folks at ISC to do so, and if you
> agree I encourage you to make your voice heard.
We hear you. Expect a decision in the next few days.
--
Evan Hunt -- ea...@isc.org
Internet Systems Consortium, Inc.
Doug Barton
>> BIND 9.6.2 is in the "b1" phase atm, which means that there is plenty
>> of time to get SHA2 in there and get the release out before a signed
>> root goes live. I encourage the folks at ISC to do so, and if you
>> agree I encourage you to make your voice heard.
>
> We hear you.
That's as much as I can hope for, thanks. :)
Mark Andrews
In message <Prayer.1.3.2.091...@hermes-1.csi.cam.ac.uk>, Chris Tho
mpson writes:
> (But it's not too obvious to me that adding support for a new signing
> algorithm should necessarily be considered a "major functional change".)
If it was *just* adding a new signing algorithm then yes it would be a minor
change. A lot more happened under the hood to support the new algorithms
on all platforms. Remember crypto support on some platforms is pretty
old and doesn't support SHA256/512 + RSA directly so we had to use more
primative methods on these platforms.
Mark