Backport DAD handling to the stable branch

[raddessi]

This backports the functionality of #232 to the stable branch so it can be used soon.

We were experiencing issues on our Debian 10 hosts similar to those in #30 where up route add ... on IPv6 interfaces would fail due to the address still being in a tentative state when the command is ran. I saw the DAD features were added in #232 but those are on the master branch which seems (for me at least) to be broken and probably in the middle of a rewrite.

I only have one outstanding issue that I can think of at the moment, shown below. Thanks!

Testing

• without enabling ipv6_dad_handling_enabled in a policyconfig file such as /etc/network/ifupdown2/policy.d/dad.json the logic operates as it does currently
• Enabling the feature flag in the above file path does enable the logic. Contents to enable the feature:

  {
      "address": {
              "module_globals": {
                      "ipv6_dad_handling_enabled": "yes"
              }
      }
  }
  

  • With no additional config, all IPv6 addresses in network config are waited on for the default period (60x 0.1s)
  • Adding the dad-attempts and/or dad-interval to a config file on any address will be picked up correctly
    • Only the first value is picked up here, you can specify it more than once and further instances are ignored. Of course only one value can be used but perhaps there is a better way to deal with this? Seems like the parent change would have this issue also so I don't think it's a big deal.
  • Any address can have dad-attempts 0 set which will forcibly bring up the address with nodad, even if dad-attempts is set somewhere else

[sohorx]

Hi, I saw your backport. Overall, I don't see any issues but I dropped some code-wise nitpicking.

[raddessi]

No big deal @sohorx I didn't want to drift from the upstream code unless I had a reason to but that is only a minor problem. All 3 are merged.