Rejo Zenger wrote:
Postfix is not configured in any strange way here, and you're right; as far as I'm aware it does indeed not act like this when an rbl host is down (which is something I haven't seen much anyway), of course this raises the question again to which extend 3th parties influence the behaviour of mailservers. I'm pretty interested in what happened at securitysage (quite an established blacklist if I'm not mistaken), if it wasn't really "down" and was blacklisting all clients that would make this a bigger issue from my point of view.++ 15/03/07 20:34 +0100 - Robert Hensel (Hensel Hosting):http://wiki.openrbl.org/wiki/Blackhole.securitysage.com). Now it seems weird that if the rbl host is unreachable Postfix decides to simply take that as a "ah well, just block everything then", or maybe something else was going on?As far as I know, this is not default behaviour of Postfix. In other words, Postfix does not behave like this or you may have configured Postfix to behave like this.
Since it is clearly an issue that is not related to a specific host, since multiple systems (inbound and servers) had the same problem I do not find it usefull or necessary to disclose that information. Also see the email below I received from securitysage (can also be found on the wiki link):Mar 14 11:01:03 hostname postfix/smtpd: NOQUEUE: reject: RCPT>from hostname[ip]: 554 5.7.1 Service unavailable; Client host[hostname] blocked using blackhole.securitysage.com; from=<email@example.com> to=<firstname.lastname@example.org> proto=ESMTP helo=<hostname>It is quite useless to paste (sections of) logfiles, while munging most of it. I could have come up with this line myself. The most important part has been munged as well: the connecting IP address which has been checked against the blackhole.securitysage.com rbl.