Roller Network

Goscomb, via HE, same route as post of 23 April: 74045 packets transmitted, 73930 packets received, 0% packet loss round-trip min/avg/max = 187.926/198.239/1548.788 ms On HE: 74124 packets transmitted, 73942 received, 0% packet loss, time 74122555ms rtt min/avg/max/mdev = 42.680/52.937/1339.736/56.1...

I've been trying to explain to HE what the problem is, but it seems to have magically gone away as of today. Mind trying it again? Hello, sorry for the delay - I didn't get a reply notification for whatever reason. From the box on Goscomb, which is traceroute'ing through HE, running this evening: 1...

Yes, looking healthier. Small loss from my box using he.net (->tiscali->sprintv6): --- 2620:0:950::242:135 ping statistics --- 7922 packets transmitted, 7836 received, 1% packet loss, time 7920924ms rtt min/avg/max/mdev = 204.014/227.875/4609.261/78.206 ms, pipe 5 and just about perfect from Goscomb...

traceroute6 and ping6 from my home box, connected via Goscomb Tech's SixXS tunnel: pooh:~ tgarcia$ traceroute6 2620:0:950::242:135 traceroute6 to 2620:0:950::242:135 (2620:0:950::242:135) from 2a01:348:165::217:f2ff:fedb:1437, 30 hops max, 12 byte packets 1 nurse.felin.es 1.046 ms 0.467 ms 0.222 ms ...

Appreciate the response. I was looking for secondary nameservers; with the recent IPv6-ifying of some root nameservers, IPv4-free DNS resolution is now feasible. For me it's just a learning (and implicit IPv6 advocacy ) exercise.

Hiya, couple of quick questions:

1. Do you have any nameservers with IPv6 addresses? (i.e. accessible via IPv6 rather than just serving AAAA / ip6.arpa PTR records)

2. Is it considered bad form to use a free account for secondary DNS but then stealth the primary?

Thanks.

As in the topic, Zone Edit for one will not send NOTIFY anywhere, so it would be nice to be able to hit something or use a forthcoming API call to force the reload... rate limited to prevent abuse if necessary I guess. Cheers.

Ok, it's making sense now: spf has a dual role and now I know about the (weirder) HELO one. Thanks for your input. BTW I think it's "mx" only rather than "mx:all" if the intention is to accept all mx for rollernet.us, the keyword "all" being solely for the ~all/-all cat...

In this case what I think I mean (I am learning) by the standard fallback thing is checking the helo/ehlo, as you mention... in other words, because mxin.mxes.net sees "helo mail.rollernet.us", it checks rollernet.us spf records. Strictly, the draft says: At least the "MAIL FROM"...

Full headers (w's, x's, y's substituted for actual usernames; remove zzzzz's entirely) given below. The headers seem to show sender @yahoo.com throughout, so I guess it's really just making the (wrong) check as if from @rollernet.us? [edit] I have just received a reply from Tuffmail technical suppor...

When trying to send mail from a yahoo.com to a hivemind.org address, and the mail has been held on mail.rollernet.us before being forwarded to primary MX (mxin.mxes.net - tuffmail imap service), I get a warning in the header linking to: http://spf.pobox.com/why.html?sender=rollernet.us&ip=208.11...