* Julien Vehent <jul...@linuxwall.info>:
> Hi Folks,
> 
> Like others before me, I'm curious about how well postscreen
> performs in the real world.
> 
> I'm writing an article on Postscreen for the french issue of linux
> magazine. I got the setup part covered fine, but miss some
> statistics on how it runs.
> I have logs from two mid-size organizations I run, but the logs
> contains only a couple thousands different IPs, so I'm wondering if
> the community would be willing to share some data :)

>From mail.python.org. It handles all list traffic for python.org:

=== Postscreen statistics ===
7 BARE NEWLINE
15 COMMAND COUNT LIMIT
165 COMMAND PIPELINING
736 COMMAND TIME LIMIT
38678 CONNECT
24189 DNSBL
25812 HANGUP
543 NOQUEUE 450 deep protocol test reconnection
4 NOQUEUE too many connections
486 PASS NEW
3455 PASS OLD
9373 PREGREET
7206 WHITELISTED

=== Clients statistics ===
3 avg. dnsbl rank
12301 clients
309 reconnections
4050 seconds avg. reco. delay

=== First reconnection delay (graylist) ===
delay| <10s   |>10to30s| >30to1m| >1to5m | >5to30m|>30mto2h| 
>2hto5h|>5hto12h|>12to24h| >24h   |
count|4       |18      |4       |83      |110     |46      |18      |17      |6 
      |3       |
   % |1.3     |5.8     |1.3     |27      |36      |15      |5.8     |5.5     
|1.9     |0.97    |


> 
> There is a script to do the log parsing:
> https://github.com/jvehent/Postscreen-Stats
> Any short report (without the IPs) would be interesting. If,
> additionally, you can calculate the geo-localisation, that would be
> great (but querying the database on hostip.info takes a while).
> 
> It's has simple as: $ python postscreen_stats.py -f maillog.1 -r
> short -g
> 
> I'm also open to suggestions on how to improve the script. I'm no
> python expert, and there are probably tons of interesting
> information to extract from the postscreen logs.
> 
> Thanks a lot,
> 
> Julien
> 
> 
> 
> -- 
> Julien Vehent - http://1nw.eu/!j

-- 
All technical questions asked privately will be automatically answered on the
list and archived for public access unless privacy is explicitely required and
justified.

saslfinger (debugging SMTP AUTH):
<http://postfix.state-of-mind.de/patrick.koetter/saslfinger/>

Reply via email to