This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Postgrey anyone?

Hi Guys, just quick question on if anyone is using postgrey in conjuction with Puremessage.   Not sure if this would make puremessage more effective.  We are on 5.5.6 and still seeing a lot of spam.  Yes I know we are advised to upgrade to 5.6.0 but we don't see the need since the spam engine has not been upgraded.  And yes we are aware that this would not be supported by Sophos.

Thanks

:5378


This thread was automatically locked due to age.
  • Hi mrdky,

    I'm not using this, and after reading the white paper on greylisting, I doubt I ever would. I have time-based constraints that I have to mangae and delaying a message by defult for 60 min isn't an option.

    Having said that, it does seem like an interesting way to help deal with some spam.

    I use the IP Block list from Sophos and this seems to cut down on the vast majority of junk for us. The rest of it is handled by the rules or custom black-lists.

    Still, I wonder if it's something Sophos might consider adding into PureMessage?

    Erric

    :5401
    • Thank's Erric for response.  We are using IP blocker as well but still getting some junk.   Wanted to get feedback on other methodologies to getting a higher catch rate.

      :5570
      • Check out http://code.google.com/p/gross/ It allows you to use multiple DNSBLs as well as the Sophos IP Blocker in a weighted algorithm for greylisting and blacklisting email. Essentially, if the sender IP is only listed on one or two blacklists then you greylist the message, and if the sender IP is on lots of blacklists then you refuse the message. The nice thing is that you can use very aggressive blacklists since false positives would only be subjected to the greylisting delay.
        :5571
        • After looking at this again, I am a little unsure of how this would be implemented since postgrey would listen to port 6000. and it looks like blockrd listens on port 4466.  I know this is off topic of support but if we put something like this in the postfix main.cf

          smtpd_client_restrictions = ignore_policy_error,

          check_policy_service inet:localhost:4466,

          check_policy_service inet:localhost:60000

          Would this work? In my mind it would go through blockrd then postgrey?

          Thanks

          :6573