I see that Mikrotik have allocated IPv6 addresses to their web server (www.mikrotik.com → 2a02:610:7501:1000::2), forum (forum.mikrotik.com → 2a02:610:7501:1000::201) and primary MX (mailgw.mikrotik.com → 2a02:610:7501:1000::199).
Unfortunately, none of the addresses in question are responding to the appropriate services. Yes, of course, browsers will fail back to IPv4, but only after a significant wait.
Basically, this means that, to all intents and purposes, Mikrotik doesn’t exist when the end user has an IPv6 connection.
$ traceroute6 www.mikrotik.com
traceroute to www.mikrotik.com (2a02:610:7501:1000::2) from 2001:8b0:fff1:0:20e:2eff:fe6d:ba68, 30 hops max, 24 byte packets
1 0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.1.f.f.f.0.b.8.0.1.0.0.2.ip6.arpa (2001:8b0:fff1::) 0.447 ms 0.339 ms 0.203 ms
2 a.gormless.thn.aaisp.net.uk (2001:8b0:0:53:203:97ff:fe05:8000) 19.999 ms 20.984 ms 20.822 ms
3 2001:7f8:4::50e8:1 (2001:7f8:4::50e8:1) 19.673 ms 20.702 ms 19.649 ms
4 linx-224.retn.net (2001:7f8:4::232a:1) 20.632 ms 20.652 ms 20.61 ms
5 rt.tnr.hki.fi.retn.net (2a02:2d8::57f5:e096) 53.89 ms 144.576 ms 53.243 ms
6 * * *
7 * * *
8 * * *
Seems like it would be worth talking to your Upstream Normis, You have 2 networks HE.net and retn.net both unable to get near your netblock. Normally 1 network I blame the network, 2 networks and I blame the network I’m trying to get to
5 2001:1900:4:3::c1 99.017 ms 98.238 ms
6 2001:1900:15:7::3 100.016 ms 2001:1900:15:6::3 98.945 ms
7 2001:1900:6:1::15 189.232 ms 188.421 ms
8 2001:1900:5:2:2::19e 203.984 ms 203.010 ms
9 RT.TNR.HKI.FI.retn.net (2a02:2d8::57f5:e096) 226.078 ms 225.320 ms
10 2a02:610:ffff:1000::1 235.545 ms 233.019 ms
11 2a02:610:ffff:1003::2 233.612 ms 232.969 ms
12 2a02:610:7500::1 233.299 ms 231.066 ms
13 2a02:610:7500:8::2 233.593 ms 231.367 ms
14 2a02:610:7500:9::2 234.031 ms 233.366 ms
15 2a02:610:7501:1000::1 295.983 ms 274.265 ms
16 2a02:610:7501:1000::2 296.917 ms 269.813 ms
Fun problem Still looks like an issue with MT v6 transit provider, It could be that it’s not getting a full route table or not getting the route for this guys connection
They told us that IPv6 is still in testing stage, and we shouldn’t be surprised that stuff breaks
That’s fine, but then you shouldn’t have created AAAA records for your existing FQDNs - perhaps 6.www.mikrotik.com and 6.forum.mikrotik.com (or whatever) should have been used until testing’s complete?
C:\Documents and Settings\ricky>tracert mikrotik.com
Tracing route to mikrotik.com [2a02:610:7501:1000::2]
over a maximum of 30 hops:
1 <1 ms <1 ms <1 ms 2001:8c8:0:xxxx::x
2 1 ms <1 ms 1 ms 2001:8c8:0:xxxx:> > x
3 23 ms 22 ms 22 ms 2001:978:2:42::25
4 * * * Request timed out.
5 * * * Request timed out.
6 * * * Request timed out.
7 121 ms 121 ms 121 ms te-11-1-0.edge2.Washington4.Level3.net [2001:1900:4:3::c1]
8 121 ms 121 ms 121 ms vl-70.car1.Washington1.Level3.net [2001:1900:15:6::3]
9 245 ms 226 ms 209 ms vl-4086.edge3.Frankfurt1.Level3.net [2001:1900:6:1::15]
10 117 ms 118 ms 117 ms 2001:1900:5:2:2::19e
11 * 181 ms 151 ms RT.TNR.HKI.FI.retn.net [2a02:2d8::57f5:e096]
12 * * * Request timed out.
13 * * * Request timed out.
14 * * * Request timed out.
15 * * * Request timed out.
16 ^C
via HE tunnel:
C:\Documents and Settings\ricky>tracert mikrotik.com
Tracing route to mikrotik.com [2a02:610:7501:1000::2]
over a maximum of 30 hops:
1 <1 ms <1 ms <1 ms 2001:8c8:0:xxxx::x
2 1 ms 1 ms 1 ms 2001:8c8:0:xxxx:> > x
3 26 ms 26 ms 26 ms tunnel169.tserv18.fra1.ipv6.he.net [2001:470:15:a9::1]
4 26 ms 26 ms 26 ms gige-g2-5.core1.fra1.he.net [2001:470:0:a5::1]
5 * * * Request timed out.
6 * * * Request timed out.
7 * * * Request timed out.
8 * * * Request timed out.
9 * * * Request timed out.
10 * * * Request timed out.
11 ^C
hm… maybe the idea with alternative quad A records isn’t so bad, at the moment… and when ipv6 become more reliable, then create simetrical A and AAAA records…
?
> x