yahoo-eng-team team mailing list archive

[Sridhar Gaddam <sgaddam@xxxxxxxxxx>]

Yes David, this is the behavior of SLAAC + EUI-64 address assignment.

** Changed in: neutron
       Status: New => Invalid

-- 
You received this bug notification because you are a member of Yahoo!
Engineering Team, which is subscribed to neutron.
https://bugs.launchpad.net/bugs/1471343

Title:
  IPV6 DHCP ports created in neutron has strange behaviour

Status in OpenStack Neutron (virtual network service):
  Invalid

Bug description:
  Some random IPV6 seems to be selected for the DHCP port when the port
  is created.

  It's not really a bug for say but this makes the behaviour less
  predictable ...

  For instance, in devstack, by default, it will create a random 40 bit
  IPV6 global ID (to comply with RFC 4193) (d23:2460:6566::/64 in this
  example) and then will create a gateway (fd23:2460:6566::1) which is
  totally predictable.   Then, for some reasons, the DHCP port gets this
  IP: fd23:2460:6566:0:f816:3eff:fe6e:cc8e ...  For some reasons, I
  would have expected it to be fd23:2460:6566::2 ...

  I would expect the VMs to get IPs that are sequentially higher than
  the gateway...  In this case,

  VM                       : fd23:2460:6566:0:f816:3eff:fe50:70ac
  DHCP/DNS/etc: fd23:2460:6566:0:f816:3eff:fe6e:cc8e

To manage notifications about this bug go to:
https://bugs.launchpad.net/neutron/+bug/1471343/+subscriptions