dhcpcd-discuss

On 28/09/2017 14:17, Graham Breed wrote:
> So I'm still mystified in that both implementations think there's
> something wrong, but I don't know what it is.

From what've been said, I don't see anything wrong.

Maybe start over with a clean slate.

I would recommend this course of action:

 * add `debug` to dhcpcd.conf so it's plenty verbose about what and why
 * add `logfile /var/log/dhcpcd.log` to dhcpcd.conf incase capturing
   console output is hard
 * start dhcpcd. I generally do `dhcpcd -dB` to force debugging and not
   forking to the background so I can monitor console output
 * one finished, post either the log file or console output here
   and state what you think is wrong

>> dhcpcd is the only all-in-one solution :)
> 
> Indeed, and we find it useful to use it like that to control all dynamic
> address and route assignment.  And it makes this all the more
> perplexing, because if dhcpcd can use the prefix length to assign a
> stateless address, why not use the same prefix length for the DHCPv6
> address?

Because there is no direct relationship between an address received by
DHCPv6 and the Prefix Route information received by Router Advertisement.
Infact the **only** relationship between DHCPv6 and Router Advertisement
is the RA saying "Start DHCPv6 and obtain an address" (this is the M for
Managed flag) or saying "Start DHCPv6 and obtain extra information but
don't obtain an address" (this is the O or Other flag).

> It wouldn't attempt to get the address at all if it hadn't
> received the correct advertisement.

Not at all.
Routing could be configured by hand, and there are other address types
such as IA_PD/Prefix Delegation which *can* install routes (but this
requires manual setup in dhcpcd).

Infact this is a very good design decision.
Say that you have Routers A and B along with DHCPv6 server D.
The host receives an RA from A and and address from D.
Then the host receives the same prefixes as in A but from B.
A announces it's no longer a router (OR the RA simply expires)

IF the DHCPv6 addresses from D were in any way tied to A then they would
have expired as well even though B is carrying the same information.

But because thankfully they are not then they stick around and still
work perfectly ok with B.

>> dhcpcd is saying that a 3rd party deleted the address.
> 
> In that case, I know what the 3rd party is, so never mind this detail.

Good, one less thing to worry about :)

Roy