I-mscp 1.1.5 antispam policy too aggressive?

    I have updated all my i-mscp servers to the latest 1.1.5 version, after this, I have notice that some of the incoming emails does not reach the inbox destination. The reason, mostly, according to the log,



    Quote

    Recipient address rejected:
    [cached] Mail appeared to be SPAM or forged. Ask your
    Mail/DNS-Administrator to correct HELO and DNS MX settings or to
    get removed from DNSBLs - retrying too fast. penalty: 30 seconds x
    0 retries.;


    Are there any changes in the postgrey/policyd-weight settings, with this new version?


    PS: A bit offtopic. Nice that i-mscp website is using piwik for analytics. ;)

    Same issue but with 1.1.3. I checked rejected IPs and domains against all blacklist and check HELO and all test result in valid servers. However, policyd-weight reject servers. May be some change in postfix package?


    SaludOS/2

    If you want to add some sending mailserver on the whitelist you have to read about policyd whitelisting. This is not working out of the box. You need a hook too, to prevent the ovewriting while upgrading i-MSCP.

    Manually add servers in the white-list isn't a solution, let's be serious. For the moment, the fast solution was to disable postgrey and policy-weight from the configuration.
    DAX, I didn't notice any of this problem in previous versions of i-mscp. If the i-mscp upgrade got nothing to do, I guess is just a pure coincidence.

    We had the problem of not receiving any emails anymore after upgrading to ubuntu 14.04 and i-mscp 1.1.5 (not sure which update was responsible)


    The error log said


    After uncommenting policy weight check in etc/postfix/main.cf the problem was gone and receiving emails is working fine.
    Any idea what the problem could be? I can live with a disabled policy weight service as long as I am able to receive my emails
    but nevertheless I am interested in knowing the reasons and another solution if there is any.

    i-MSCP 1.5.3 | Ubuntu 18.04

    Edit the /etc/policyd-weight.conf, find 'rbl.ipv6-world.net', 4.25, 0, 'IPv6_RBL' and uncommenting this. This was the problem of my Server.

    What entries do you mean?
    We have uncommented
    # check_policy_service inet:OUR.IP.XX.YXZ:12525,
    # check_policy_service inet:127.0.0.1:10023,
    in etc/postfix/main.cf


    and the problem is gone now. Prior to this we didn't receive any mail from outside.
    Not only those that might be considered spam but not a single one ...
    This is what I finde strange.

    i-MSCP 1.5.3 | Ubuntu 18.04

    My apologies for bringing this thread back up, but I am having similar problems:


    Code
    1. Jul 18 03:23:50 ginger postfix/policyd-weight[30287]: child: spawned
    2. Jul 18 03:23:50 ginger postfix/policyd-weight[30287]: warning: child: could not open RBL Lookup Socket to config: IO::Socket::INET: Bad hostname 'config' Invalid argument
    3. Jul 18 03:23:50 ginger postfix/policyd-weight[30287]: warning: child: err: can't resolve "config" to address at /usr/lib/perl5/Net/DNS/Resolver/Base.pm line 755, <GEN155> line 26.#012
    4. Jul 18 03:23:50 ginger postfix/policyd-weight[26839]: master: child 30287 exited


    Adding warn_if_reject to main.cf makes my server accept mails again, but just as bluecafe I would like to find the actual problem here. I have done a fair bit of internet search, but most similar errors are because of wrong ports being used or deamons not listening. This looks different, but I don't know where to start checking regarding this error, especially since I have no idea why policyd-weight thinks "config" could be a hostname to connect to. Any idea where I should start looking?


    Edit: For me the problems started after I run a regular update, including policyd-weight:all 0.1.15.2-5 => 0.1.15.2-5+wheezy1 and did not go away even after I updated to i-mscp 1.1.12 (but of course I still don't know whether the problem is related to the policyd-weight update or something else).