On Sun, Aug 23, 2015 at 2:36 PM, Elia Pinto <gitter.spiros@xxxxxxxxx> wrote: > Il 23/Ago/2015 20:26, "Eric Sunshine" <sunshine@xxxxxxxxxxxxxx> ha scritto: >> On Sun, Aug 23, 2015 at 1:16 PM, Johannes Löthberg >> <johannes@xxxxxxxxxxxx> wrote: >> > Just an A record would be enough. The issue is that mail.sunshineco.com >> > has >> > neither an A nor an AAAA record, it is a CNAME to sunshineco.com, which >> > is >> > invalid according to RFC2181. >> >> Interestingly, the default configuration for all domains managed by >> this service provider is for the mailhost to be a CNAME. While the >> restriction in section 10.3 of RFC2181 makes sense as a way to avoid >> extra "network burden", in practice, email services seem to be pretty >> relaxed about it, and follow the CNAME indirection as needed. >> >> I suppose it's possible that web.de is being extra strict (although it >> seems that such strictness would be painful for its users), or this >> could just be a temporary DNS lookup failure. It's hard to tell based >> upon the errors René reported. >> >> I did change the CNAME to an A just in case, though who knows how long >> it will take for the change to propagate over to web.de's server. > Anyone can check Here https://dnschecker.org/#CNAME/Mail.sunshineco.com > It would fail with your change Interesting service; thanks for the pointer. However, since it's just querying a random set of DNS servers, it's not necessarily indicative of whether the change has actually propagated to the DNS server(s) answering web.de's mail server's queries. Local configuration (TTL's, etc.) on those servers or anywhere in between, as well as network conditions, could impact propagation to an unknown degree. -- To unsubscribe from this list: send the line "unsubscribe git" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html