IPv6. It's gonna happen. It *has* to happen.

These are facts.

WTF to do about it ? Ignore it ?

I guess the 'IPv6' menu option will be the answer.

However, unless all the stuff we need works with IPv6, then *we're* screwed, which in turn means we buy no more Mikrotiks.

Beta will be Fine - it will all be Beta for quite some time - IPv6 is itself still Beta.

The IPv4 to IPv6 changeover will be a step-change, not a smooth transition.

Make it so MT does IPv6 all over and you'll win big-time.

the RB433 I have on my desk for abuse has ipv6 package installed but disabled, I'm not sure if thats included in the combined package or not. But it is at least available in the all packages download option for 5.0rc3

IPv6. It's gonna happen. It *has* to happen.

These are facts.

WTF to do about it ? Ignore it ?

I guess the 'IPv6' menu option will be the answer.

However, unless all the stuff we need works with IPv6, then *we're* screwed, which in turn means we buy no more Mikrotiks.

Beta will be Fine - it will all be Beta for quite some time - IPv6 is itself still Beta.

The IPv4 to IPv6 changeover will be a step-change, not a smooth transition.

Make it so MT does IPv6 all over and you'll win big-time.

Assuming this is a serious post from you about v6 then here's some help

First off IPv6 isn't in beta, It's been a standard for over 10 years, DHCPv6 has been around for 7 years etc. MT's currently implementation is patchy, Core things like routing and address's are stable in v4, Things needed for dynamic clients like DHCPv6 and PPPoEv6 are only in v5 beta. Accessing a MT box over v6 is only telnet at the moment, Winbox,SSH and API are coming in v5. All other tools like bandwidth test etc are not working and there is also no clue on when MPLS will be sorted (mpls really can stay on private v4 for the time being)

As for v4 to v6 transition, It can be smooth if done right, You can use dual-stack which is RFC1918 v4 space on client lan and same again on client WAN then NAT it at your ISP/Upstream connection to public v4. On top of this you run your v6 stack using public v6 to the client. NAT64 down the line and/or there is bound to be an off the shelf product for this in the coming months/years.

Most RIR's are handing out /32 v6 space to existing holders of v4 space and for initial testing/low load you can run a bgp tunnel over 6to4 to he.net for you're v6 connection and your away laughing

unless all the stuff we need works with IPv6

nobody knows all the stuff you need, so be more specific

Click on the top option in Winbox, click all the sub-tabs.

Carry on clicking until you reach the bottom.

All that stuff - the stuff in ROS that we buy ROS for.

I hate to say it but he's right, 12 months from now we're going to need most if not all features in day to day running of networks in v6, We've got PPPoE in beta along with SSH access and some tools, Next we'll need things like Winbox and Btest, DHCPv6 for those networks where RA wont cut it, Some of the basic's like NTP server/client, Radius etc.

Depending on the model you follow we're nearing the 100-days-till-IANA-empty, 184million left, When we hit 84million left tho IANA hands out 16million to each RIR and thats it (Going to suck for APNIC) So the reality is some of us have 3 months to see if ROS 5 is stable enough for mass deployment with the v6 features we need

Two /8s allocated to APNIC from IANA
https://www.apnic.net/publications/news/2011/delegation

Published on: Tuesday, 1 February 2011

Dear Colleagues

The information in this announcement is to enable the Internet community to update network configurations, such as routing filters, where required.

APNIC received the following IPv4 address blocks from IANA in February 2011 and will be making allocations from these ranges in the near future:

•39/8
•106/8
Reachability and routability testing of the new prefixes will commence soon. The daily report will be published on the RIPE NCC Routing Information Service.

Please be aware, this will be the final allocation made by IANA under the current framework and will trigger the final distribution of five /8 blocks, one to each RIR under the agreed "Global policy for the allocation of the remaining IPv4 address space".

After these final allocations, each RIR will continue to make allocations according to their own established policies.

APNIC expects normal allocations to continue for a further three to six months. After this time, APNIC will continue to make small allocations from the last /8 block, guided by section 9.10 in "Policies for IPv4 address space management in the Asia Pacific region". This policy ensures that IPv4 address space is available for IPv6 transition.

It is expected that these allocations will continue for at least another five years.

APNIC reiterates that IPv6 is the only means available for the sustained ongoing growth of the Internet, and urges all Members of the Internet industry to move quickly towards its deployment.

So Normis, Where are all those "We've heard this before" post's of yours?

The first seal has been broken on v4 Hell, IANA is out. APNIC will be out by year end *AT MOST*, Now's MT's time to step up and push out some production quality v6 software in ROS 5.0 quickly, And I mean production quality. We're sitting here waiting to deploy v6 because BGP v6 in ROS 4.16 is unstable. APNIC realistically has 3-5 months (They burned thru 2 /8's in 3 months, Increase demand and I personally expect to see them run out in 3 months flat)

Everyone's watching now

Last time we had this discussion, january and february 2011 were mentioned.

But you've seen our ToDo list, so don't worry.

I've seen your todo list, It's great. My point of my post "PRODUCTION QUALITY" which is somewhat lacking considering v6 BGP is broken in ROS 4 and wont be looked into and ROS 5 is the autistic step child rocking in the closet i.e Quirky.

The feature's your putting in are great but you have to deliver software that can go into a production setting, History teach's us that ROS releases take till around .7-.10 release to become really stable and we're still sitting on RC7 month's away from RIR depletion. We're ready to deploy v6 network wide right now, BGP v6 stability in ROS 4.16 is stopping us, Here's an entire deployment of v6 which has worked around all of ROS 4's v6 lacking stuck because dynamic routing is broken in ROS 4

Release speed is not related to laziness, there is no way you can "convince" us to make stable software 3x faster. We are doing what we can, and development doesn't stop for a minute here. We have a list of programmers working hard all day, every day.

Last time we had this discussion, january and february 2011 were mentioned.

But you've seen our ToDo list, so don't worry.

I hope that DHCPv6 is high up on the to do list.

We're ready to deploy v6 network wide right now, BGP v6 stability in ROS 4.16 is stopping us, Here's an entire deployment of v6 which has worked around all of ROS 4's v6 lacking stuck because dynamic routing is broken in ROS 4

Whats not working with bgp v6 ? I hope you have opened tickets and sent supouts, otherwise they will say it works fine and no one is complaining : ) I am on 3.30 using BGPv6 for a year now without any problems, up until this week, and now I have major problem with Cogent peer, it wont stay up once it receives all routes. Cant even ping the neighbor IP. Was thinking of going to 4.16 but not if its just as bad. Can you elaborate?

you always can go to 4.16 or even 5.0rc8 and then go back to version you are using now.

We're ready to deploy v6 network wide right now, BGP v6 stability in ROS 4.16 is stopping us, Here's an entire deployment of v6 which has worked around all of ROS 4's v6 lacking stuck because dynamic routing is broken in ROS 4

Whats not working with bgp v6 ? I hope you have opened tickets and sent supouts, otherwise they will say it works fine and no one is complaining : ) I am on 3.30 using BGPv6 for a year now without any problems, up until this week, and now I have major problem with Cogent peer, it wont stay up once it receives all routes. Cant even ping the neighbor IP. Was thinking of going to 4.16 but not if its just as bad. Can you elaborate?

After a while it stops routing, the routes sit in the route table but attempting to reach any of the hosts results in "no route to host", Restart BGP peer and it's fine.

MT's response was that v6 dynamic routing was buggy and wouldn't be fixed until ROS 5.0

4.16 has the problem and once *AGAIN* we run a production network, we are not going to load RC's onto core routers which is where we need them since we are ready to roll v6 wise

do you have more than 1 ipv6 bgp peers?

Of course :) We have 4 peers. We only have to restart the peer affected during the issue. Doesn't happen with v4 in the same setup

Yes, I have 4 peers. It normally affects 1 peer and a simple peer disable/enable clears up the issue. Sometimes the issue only affects certain routes and these routes are both overlapping with other peers and also the only route for it.