L
lnxgeek

  • Sep 16, 2024
  • Joined Aug 11, 2023
  • 8 discussions
  • 24 posts
  • 3 best answers
  • Post posted... wait what? You got the answer!
  • After I deleted the mentioned files the problem went away for me.

    You could try the same as a first step.

  • Roger 👍

    After cruising through the webui, I found that I could change the score in the dedicated Rspamd UI.
    We will see if it takes effect :-)

  • Hi
    If I want to change the score for MISSING_TO in Rspamd where is the best place to make this change?

    I can see that I could add it in the “/data/conf/rspamd/local.d/composites.conf” but will it be replaced after an upgrade?

    • No, /data/conf is for all persistant settings and should survive an upgrade.

  • Hehe :-)

    The problem is that, for me at least, when it is left it is a bright color in dark mode, and in other interfaces right is off. It helps a bit in light mode.

    Also, there no other toggles to reference from (except dark mode toggle), so I couldn’t find a way to get a visual cue.

    But thanks for helping out, much appreciated.

  • Hi
    With the 2023-12 release a new toggle has turned up under Fail2ban parameters.

    I find it hard to read, if it is turned on or off and there are now other visual cues to indicate the state.

    So, can someone help an old man read the toggle, when is it on and when is it off?

    • Left = deactivated (greyed out, white)
      Right = activated (solid blue)

      I am also almost 60, but have no problem to read the state.
      When have you been to the oculist lately? 😉

  • I’ve tried the solution with the Docker daemons and you need to disable docker.sock as well:

    systemctl disable docker.service docker.socket

    to keep it from starting up.

  • lnxgeek
    It sould seem that it is related to some extractors, so it isn’t Dovecot related.

  • Hi all

    I want to consume the logs into Graylog but I can see that the time format is “%b %d %H:%M:%S ” when it arrives.
    When looking in dovecot.conf I see that it should be in the format “%Y-%m-%d %H:%M:%S”.

    My logs are sent vi docker (docker-compose.override.yml):
    dovecot-mailcow: # or any other
    logging:
    driver: “syslog”
    options:
    syslog-address: “tcp+tls://syslog.example.com:514”
    syslog-tls-ca-cert: “/etc/ssl/ca.pem”
    tag: “dovecot-mailcow”
    syslog-format: “rfc5424”

    Don’t know if this has something to do with it.

    • lnxgeek replied to this.
    • lnxgeek
      It sould seem that it is related to some extractors, so it isn’t Dovecot related.

    • Dammit I have googled like crazy and never seen that one.

      Thanks!

    • Hey everyone

      Does anyone know what this part of the rbl line in postfix means: "*-2" or "*2"

      I know what the other part of the equation means but asterisk is a puzzle to me.

      From main.cf:
      postscreen_dnsbl_sites = wl.mailspike.net=127.0.0.[18;19;20]-2
      hostkarma.junkemailfilter.com=127.0.0.1
      -2
      list.dnswl.org=127.0.[0..255].0-2
      list.dnswl.org=127.0.[0..255].1
      -4
      list.dnswl.org=127.0.[0..255].2-6
      list.dnswl.org=127.0.[0..255].3
      -8
      ix.dnsbl.manitu.net2
      bl.spamcop.net
      2
      bl.suomispam.net2
      hostkarma.junkemailfilter.com=127.0.0.2
      3
      hostkarma.junkemailfilter.com=127.0.0.42
      hostkarma.junkemailfilter.com=127.0.1.2
      1
      backscatter.spameatingmonkey.net2
      bl.ipv6.spameatingmonkey.net
      2
      bl.spameatingmonkey.net2
      b.barracudacentral.org=127.0.0.2
      7
      bl.mailspike.net=127.0.0.25
      bl.mailspike.net=127.0.0.[10;11;12]
      4
      dnsbl.sorbs.net=127.0.0.108
      dnsbl.sorbs.net=127.0.0.5
      6
      dnsbl.sorbs.net=127.0.0.73
      dnsbl.sorbs.net=127.0.0.8
      2
      dnsbl.sorbs.net=127.0.0.62.23

    • But of course 😆
      Thanks for yet another great input 👍

    • What happens if you just ssh between the servers?

      Have you tried to just rsync’ing random folder across manually?

    • Just to follow up on this one.
      I created an encrypted file with cryptsetup (LUKS) and mounted it onto “mailcowdockerized_crypt-vol-1/_data”.
      Moved the files from the original “mailcowdockerized_crypt-vol-1” and into the now encrypted one and it just works :-)

      All this has of course been done with all containers shutdown.

      It was crazy easy which just show how well designed and packaged mailcow is.
      The only caveat is that you need to modify the docker-compose.yml and change all “restart: always” to “restart: no”. This stops the services from starting after boot. As you need to manually mount the volume before dovecot can read the keys, every reboot requires manual intervention. This is a little price to pay for protection of the mails.

      I have injected a short curl command in the crontab which pings me constantly with Pushover messages until I react:
      curl -s -X POST -d token=APP_TOKEN&user=USER_TOKEN&message=Server_rebooted&title=Reboot&priority=2 https://api.pushover.net/1/messages.json

      • It would seem that if I delete the corresponding dovecot.* files in “mailcowdockerized_vmail-index-vol-1” “.INBOX” the problem went away.

        Time will tell if this was a one off issue.

      • Hi
        I am starting to see a lot of these errors on a user:

        imap(xxx@example.com)<117976><1kR5ozIDwFxXNjqC>: Error: Mailbox INBOX: UID=1726: read() failed: Cached message size larger than expected (7720 > 4096, box=INBOX, UID=1726) (FETCH BODY[])

        Any good ideas as to how to fix this?

        • It would seem that if I delete the corresponding dovecot.* files in “mailcowdockerized_vmail-index-vol-1” “.INBOX” the problem went away.

          Time will tell if this was a one off issue.

      • I had that idea but it is not easy to implement on a VPS.

        I like the construction where it is part of the design. If I implement my own design I have to maintain it when there are updates and such.
        My reasoning for using mailcow is its beautiful way of collecting all the different pieces into one manageable solution. So, building my own kinda go against it all :-)

      • I love the idea that Dovecot default encrypt the mail files at rest on the disk.

        But is it possible to protect the encryption files in such a way that someone with access to the disk can’t just decrypt the files?
        I know it will require a prompt to enter a password and it will make a server reboot a manual process, but it will be really awesome add this layer of protection and make it just a bit harder to get to the mails.

        • Just to follow up on this one.
          I created an encrypted file with cryptsetup (LUKS) and mounted it onto “mailcowdockerized_crypt-vol-1/_data”.
          Moved the files from the original “mailcowdockerized_crypt-vol-1” and into the now encrypted one and it just works :-)

          All this has of course been done with all containers shutdown.

          It was crazy easy which just show how well designed and packaged mailcow is.
          The only caveat is that you need to modify the docker-compose.yml and change all “restart: always” to “restart: no”. This stops the services from starting after boot. As you need to manually mount the volume before dovecot can read the keys, every reboot requires manual intervention. This is a little price to pay for protection of the mails.

          I have injected a short curl command in the crontab which pings me constantly with Pushover messages until I react:
          curl -s -X POST -d token=APP_TOKEN&user=USER_TOKEN&message=Server_rebooted&title=Reboot&priority=2 https://api.pushover.net/1/messages.json

      • mlcwuser Thanks.
        Blocking the ports was my last resort as I like to disable stuff I don’t want.

        But thanks for the input anyway :-)