dealing with Yahoo slowness

*now* specifically for dealing with Yahoo. Thanks. -- Florin Andrei http://florin.myip.org/

Re: dealing with Yahoo slowness

it but I'm not sure how big of a difference they would make. -- Florin Andrei http://florin.myip.org/

Re: dealing with Yahoo slowness

On 06/14/2010 11:13 AM, Wietse Venema wrote: Florin Andrei: P.S.: We're using postfix-2.3.3-2.1.el5_2 that comes with Red Hat 5. I'm That is two Postfix versions before _rate_delay was introduced. You may want to upgrade to Postfix 2.5 or later. Aw great. :( Sometimes Red Hat'

Re: dealing with Yahoo slowness

On 06/14/2010 11:54 AM, Florin Andrei wrote: Well, that does it. I got RPM packages with 2.7 from two different sources. Time for testing, then upgrade, and I'll keep y'all posted with the results. And here it is, the status update. I got the 2.7.0 src.rpm packages made by Simon J

Re: dealing with Yahoo slowness

= 1s yahoo_destination_concurrency_positive_feedback = 1/3 yahoo_destination_concurrency_negative_feedback = 1/8 I'll have another batch of emails to send pretty soon, I'm still tweaking parameters until then. -- Florin Andrei http://florin.myip.org/

Re: dealing with Yahoo slowness

On 06/21/2010 11:31 AM, Victor Duchovni wrote: On Mon, Jun 21, 2010 at 11:08:04AM -0700, Florin Andrei wrote: yahoo_destination_concurrency_limit = 4 yahoo_destination_concurrency_failed_cohort_limit = 5 yahoo_destination_rate_delay = 1s yahoo_destination_concurrency_positive_feedback = 1/3

Re: dealing with Yahoo slowness

On 06/21/2010 12:42 PM, Victor Duchovni wrote: On Mon, Jun 21, 2010 at 12:21:45PM -0700, Florin Andrei wrote: yahoo_destination_concurrency_failed_cohort_limit = 20 yahoo_destination_rate_delay = 1s I can't say I understand *why* the 1s rate delay makes the feedback and the concurrency

Re: dealing with Yahoo slowness

ith minimal_backoff_time = 1000 and maximal_backoff_time = 2000. I'll try 500 and 1000 instead, maybe that makes the blue bumps more narrow. -- Florin Andrei http://florin.myip.org/

DNS load-balancing two equal nexthops is not fair

rify how and why I end up with the 3:1 or 4:1 distribution? What makes one system receive more emails? Is it because it's more responsive? (closer topologically, also faster hardware) What's the algorithm? -- Florin Andrei http://florin.myip.org/

Re: DNS load-balancing two equal nexthops is not fair

disabling the connection cache will equalize the distribution. Or is it not that simple? Note: The systems are pretty fast and the connections are not slow either - one is local, the other is over a reasonably fast data link. -- Florin Andrei http://florin.myip.org/

Re: DNS load-balancing two equal nexthops is not fair

speed decrease might be beneficial). I think I know how to eliminate concurrency, but I'm lacking a volume-based limit for the connections. I'll keep looking for a solution. -- Florin Andrei http://florin.myip.org/

Re: DNS load-balancing two equal nexthops is not fair

xclusive constraints. Having multiple exit points seems to improve the overall delivery speed - this is true even right now, when distribution is skewed to the faster server 4:1. My estimate is, a near-1:1 distribution would actually fix our time-constraint problem even before whitelisting. So you

Re: DNS load-balancing two equal nexthops is not fair

ected. But that's ok, the internal rate is orders of magnitude above the Yahoo rate anyway. From an external perspective, things are actually much better now. Case closed. Thanks for all the help. -- Florin Andrei http://florin.myip.org/

filtering mail from outside with dynamic address

pretty convoluted. I prefer to keep things simple, hence this inquiry. -- Florin Andrei http://florin.myip.org/

Re: filtering mail from outside with dynamic address

Florin Andrei wrote: Is there a way to bind the listener to an interface using the interface name (eth5:smtp) instead of the IP (1.2.3.4:smtp)? Also, you know what would *really* help? The ability to say: "bind to all interfaces except this one", by name. That would be really, re

Re: filtering mail from outside with dynamic address

Wietse Venema wrote: Florin Andrei: Florin Andrei wrote: Is there a way to bind the listener to an interface using the interface name (eth5:smtp) instead of the IP (1.2.3.4:smtp)? No. The bind(2) system call specifies an address. Not an interface, and not the route. Connections with source

Re: filtering mail from outside with dynamic address

rity- and other-wise. -- Florin Andrei http://florin.myip.org/

real-world issues with smtpd_tls_ask_ccert?

rs are disabled when client certs are used. All the crypto stuff (CA, server cert, client cert) is ok, I tested it already with the email client and Dovecot (secure IMAP). -- Florin Andrei http://florin.myip.org/

Re: real-world issues with smtpd_tls_ask_ccert?

client is an iPhone and the mail config is rather primitive) but in the end it worked pretty well. So, now I'm not worried about that option, since the listener on port 25 is non-TLS. Thanks, -- Florin Andrei http://florin.myip.org/

anyone? success with smtpd_tls_req_ccert and iPhone as client?

tects the cert and installs it automatically. -- Florin Andrei http://florin.myip.org/

Re: anyone? success with smtpd_tls_req_ccert and iPhone as client?

PN, but it doesn't work on the iPhone, so I would have to install / configure an IPSec thing from scratch if the iPhone doesn't play nice with SMTP / SSL / SASL. It's not rocket science but it's a lot of tedious work. -- Florin Andrei http://florin.myip.org/

best way to stop all outbound delivery?

t no message should ever leave this box, for no reason, even if it's a notification for delivery error. I could block outbound port tcp/25 with iptables, but it seems inelegant. Would this do the trick? default_transport = error:no outbound emails, sorry -- Florin Andrei http://florin.myip.org/

forward the bounce message to Reply-To

field of the original message. This should apply only to bounces delivered to this particular inbox. Sounds like a procmail job, but if it's doable in Postfix alone I'd like to take that route since it's less resource-intensive. -- Florin Andrei http://florin.myip.org/

Re: forward the bounce message to Reply-To

rt in different parts of the code. I do not intend to mess with regular email. No regular email is being sent through these machines. -- Florin Andrei http://florin.myip.org/

Re: forward the bounce message to Reply-To

On 02/26/2013 01:48 PM, Wietse Venema wrote: Florin Andrei: Sending out messages through a Postfix server. Delivery is refused for whatever reason (e.g. recipient does not exist), and then a bounce is sent by Postfix to a local inbox on that server, as a failure notification. No. It is sent

allow some senders, but block everyone else

random emails appear to be blocked. I'm not sure what I'm doing wrong. -- Florin Andrei http://florin.myip.org/

Re: allow some senders, but block everyone else

hten me. Thank you. -- Florin Andrei http://florin.myip.org/

Re: allow some senders, but block everyone else

On 07/22/2013 05:30 PM, Wietse Venema wrote: Florin Andrei: The goal is to send most emails to local, send most mydomain.com recipients to a relay nearby, and let foobardomain.com senders go out on the Internet freely. Presumably, if foobardomain.com senders send mail to local or

Re: allow some senders, but block everyone else

recipient. AFAICT, the transport table will override my sender_dependent_default_transport_maps stuff no matter what. Is there a way to achieve sender-based routing before any recipient-based decision is made? -- Florin Andrei http://florin.myip.org/

Re: allow some senders, but block everyone else

search and tests are performed), but I was amazed how much easier it was to solve a complex problem like this with Exim. -- Florin Andrei http://florin.myip.org/