[ Apologies to those of you who receive this note in multiple forums. ]
Hello all,
I wanted to drop a quick note to everyone and explain how/why things
took so long and I deeply apologize for the service interruption.
Apparently the machine hosting outages mailing list went belly
up during the package upgrade (postfix/mailman/apache) in order to bring
the system up to date. This was suppose to be a planned upgrade which
unfortunately turned out to be sysadmin's nightmare. During the headless
chicken syndrome it led into further issues and from there on murphy
took over that led to a prolonged outage.
We are working towards a cluster setup (active-active cluster) where we
will be able to pull the host out of operation without affecting the
service in the future. Something we are also looking into is a separate
instance in availability zones (multi-site) in order to protect
applications/ host availability from failure of a single location.
We deeply regret the delay this caused the mailing list to be off the air.
Many thanks to Gadi Evron / Randy Vanghn / James Eastman / Larry Vaden /
Joe St Sauver and other members of the team who work effortlessly to
resurrect this list. Without there help and direction outages list
wouldn't exist today. I'm indebted to them.
Not quite. Outages list traffic appears to be originating from
209.151.108.146, which has rDNS mx2.amigostecnicos.net, which has
forward DNS 209.151.108.152, which does not match the originating IP.
This inconsistency may cause deferral/rejection/quarantining depending
on how recipients' MTAs are configured. For example, sendmail with
require_rdns switched on responds with a 451 to this.
Not quite. Outages list traffic appears to be originating from
209.151.108.146, which has rDNS mx2.amigostecnicos.net, which has
forward DNS 209.151.108.152, which does not match the originating IP.
This inconsistency may cause deferral/rejection/quarantining depending
on how recipients' MTAs are configured. For example, sendmail with
require_rdns switched on responds with a 451 to this.
I suggest fixing the DNS inconsistency.
---Rsk
- --------------------------
outages mail is being delivered.....but the second entry is botched,
we'll address it.
isotf.org has address 209.151.108.158
isotf.org mail is handled by 10 mail.isotf.org.
isotf.org mail is handled by 50 mx2.amigostecnicos.net.
vrode@promiscious:~$ host mx2.amigostecnicos.net.
mx2.amigostecnicos.net has address 209.151.108.152
vrode@promiscious:~$ host 209.151.108.152
152.108.151.209.in-addr.arpa domain name pointer mx.amigostecnicos.net.
vrode@promiscious:~$ host 209.151.108.146
146.108.151.209.in-addr.arpa domain name pointer mx2.amigostecnicos.net.
I was clearly not paying enough attention to the message to which I
replied... but why did 2 copies of it make it through the machinery?
I would have assumed nanog@nanog and nanog@merit went to the same copy
of Mailman, which should have noted the identical message-id's and
bounced one; no?
