problem with dhcp setup

[Shane Kerr]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Well... hard to say what the user wanted. Did they want a /64? Did they want a
/128? The only thing for certain is the actual declaration was broken. :)

John Jason Brzozowski (CISSP, RHCT) wrote:
> Ok so in the case originally cited the declaration could have been
> fec0:3:0:1::/64?
> 
> I think a warning is you suggest is a good idea.
> 
> 
> On 6/14/07 5:48 AM, "Shane Kerr" <Shane_Kerr at isc.org> wrote:
> 
> John,
> 
>>>> About the range6: you probably wanted to specify /128 rather than /64. We
>>>> should
>>>> probably make it an error if you specify an invalid network (like 1:2:3:4::/8
>>>> or
>>>> something). I'll try to get this fixed for 4.0.0a2 also.
>>>>> [jjmb] So a range6 using a /128 would allow for the assignment of a single
>>>>> IPv6 address?  I do not believe this should be limited to a /128, in fact,
>>>>> it should allow for variable prefix lengths.
> One problem here is that fec0:3:0::100/64 is not a valid prefix. Address
> expanded out, with a bitmask for /64 below:
> 
>     fec0:0003:0000:0000:0000:0000:0000:0100
>     1111:1111:1111:1111:0000:0000:0000:0000
>                                        ^^^
> 
> That bit on the right-hand side is invalid, which is why I suggested /128. The
> next alpha of the software will detect this error and let you know about it,
> rather than silently masking out invalid bits.
> 
> The software allows any prefix length, from /0 to /128, or any arbitrary
> range,
> for range6 statements.
> 
>>

> ===============================================
> John Jason Brzozowski (CISSP, RHCT)
> jjmb at jjmb.com
> (p) 484-994-6787
> (f) 610-616-4535
> ===============================================




-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFGcSkPMsfZxBO4kbQRAuIiAKCeYiArzZQH5jlTjEQc6sPUJqmpxwCgnDak
RVNMH7jBDHFBPnjNSoygQJo=
=d0FN
-----END PGP SIGNATURE-----