I was googling to check if there are any howtos to setup
smtp-source/smtp-sink to test smtp performance, but there are none and
after a quick dig through the man page and another article at
postfixmail.com [http://postfixmail.com/blog/?p=218] I figured that
out. I am posting the commands here for archives.
# smtp-sink -c 192.168.3.35:25 3
-c Display running counters that are updated whenever an SMTP session
ends, a QUIT command is executed, or when "." is received.
192.168.3.35:25 is the ip port where you want smtp-sink to listen.
3 is the backlog. The backlog argument defines the maximum length to
which the queue of pending connections for sockfd may grow.
If a connection request arrives when the queue is full, the client may
receive an error with an indication of ECONNREFUSED
I have no idea what would be a nice value for backlog, but this is a
required value. so I put 3. Hope others might chime in.
-v can show the SMTP conversations. -w delay can insert delay seconds
before responding to a DATA command.
This example shows 40 parallel sessions (-s 40), almost 10KB sized
messages (-l 10120), 500 messages sent (-m 500), counter display (-c),
envelope sender and receiver (-f [hidden email] -t
[hidden email]) and connection on port 25 of 192.168.3.74
-N Can prepend a non-repeating sequence number to each
recipient address. This avoids the artificial 100% hit rate in the
resolve and rewrite client caches and exercises the trivial-rewrite
daemon, better approximating Postfix performance under real-life
-r recipient_count increases the number of reciepients per mail from
the default of one.
-S subject Send mail with the named subject line (default none)
On Wed, Jun 17, 2009 at 03:44:22PM +0530, Rajkumar S wrote:
> I was googling to check if there are any howtos to setup
> smtp-source/smtp-sink to test smtp performance, but there are none and
> after a quick dig through the man page and another article at
> postfixmail.com [http://postfixmail.com/blog/?p=218] I figured that
> out. I am posting the commands here for archives.
Performance measurements are for experts. Non-experts doing performance
measurements get meaningless results. I don't see a solution, this is
not an easy problem.
If my response solves your problem, the best way to thank me is to not
send an "it worked, thanks" follow-up. If you must respond, please put
"It worked, thanks" in the "Subject" so I can delete these quickly.