Hi @all,
by accident I found out, that the SMTP Server that sends Mails for www.simplemachines.org is not properly configured.
In the SMTP Dialog it claims to be "httpd3.localdomain" - A hostname that can't apparantly be resolved. Even worse - the reverse record for its IP adress 66.71.247.155, which is fw01.atl.forceitx.net, can't be resolved, too.
Checking the Hostname in the HELO dialog against the reverse DNS record is a pretty basic method to avoid spam.
So I guess I'm not the only one who does not receice any mails from www.simplemachines.org anymore.
Attached you will find an example log
Jan 30 12:53:16 lemmy postfix/smtpd[32502]: warning: hostname fw01.atl.forceitx.net does not resolve to address 66.71.247.155: Temporary failure in name resolution
Jan 30 12:53:16 lemmy postfix/smtpd[32502]: connect from unknown[66.71.247.155]
Jan 30 12:53:16 lemmy postfix/smtpd[32502]: NOQUEUE: reject: RCPT from unknown[66.71.247.155]: 450 4.7.1 <httpd3.localdomain>: Helo command rejected: Host not found; from=<
[email protected]> to=<
[email protected]> proto=ESMTP helo=<httpd3.localdomain>
Jan 30 12:53:16 lemmy postfix/smtpd[32502]: disconnect from unknown[66.71.247.155] ehlo=1 mail=1 rcpt=0/1 data=0/1 rset=1 quit=1 commands=4/6
Could you please forward this to your Mailadmin ?
In case there are any questions left I'd be glad to help you.
Bye !
Thank you, I've moved this to a better board.
Thanks you !
"Site Comments, Issues and Concerns" fits much better, indeed. Looks like I oversaw this - Sorry for that !
Thanks for your report.
This is a known-issue that comes and goes when we have to switch the delivery method and the mismatching domains and IP's is a result of the the way the stack functions internally. It's essentially a failover method that then goes through a layer of network translation and unfortunately sometimes may run in that configuration for an extended period of time.
A permanent fix will be implemented and is on the to-do list.
Fortunately a new server that can relieve the problem has recently arrived in the datacenter, just waiting on some parts. :) I'll see if I can re-route it without issues in the meantime.
Cool, I'm glad that this issue is already known and that there are already plans to fix this.
Thanks a lot for your feedback :)
Hello,
This problem should be resolved for you now, even when in the backup mode the servers will now send email with valid hostname, PTR and SPF records.
I had missed you said the PTR was mismatching, that turned out to be an error upstream in the datacenter; I did change that a while ago, but the change never made it to RL.
I bet this is what bit me when I was trying to register initally.
Thanks for fixing it!