Posts by fulltilt

    it's an ip-set blacklist which make use of some major BL too f.ex. Spamhaus ... attacking IP addresses are set to reject and in this case the letsencrypt verification server was blocked. It's not a plugin issue but maybe this information helps when using external blacklists or fail2ban RBL rejects too.

    yep, the letsencrypt IP address is blacklisted (Spamhaus SBL-XBL Combined Block List) ... if anyone uses ip-set blacklists you should whitelist 64.78.149.164

    http://multirbl.valli.org/lookup/64.78.149.164.html


    Code
    1. 64.78.149.164 - - [04/Mar/2019:10:14:06 +0100] "GET /.well-known/acme-challenge/spx1448aEpmFztj0kBCis5VLOKgrfjuH8ty28JdYJRA HTTP/1.1" 200 87 "-" "Mozilla/5.0 (compatible; Let's Encrypt validation server; +https://www.letsencrypt.org)"
    2. 64.78.149.164 - - [04/Mar/2019:10:14:06 +0100] "GET /.well-known/acme-challenge/6TMquHMqNbH6eGIC2WUxwFkdY4t3EtYaoF9tiL3GZmM HTTP/1.1" 200 87 "-" "Mozilla/5.0 (compatible; Let's Encrypt validation server; +https://www.letsencrypt.org)"



    - IMSCP: 1.5.3

    - Distribution: Debian 9

    - Proftpd

    - PHP FPM

    - MariaDB 10.1

    - Dovecot

    - Roundcube

    - Web2FTP

    - Plugins:

    PMA Captcha, RoundCubePlugins, SpamAssassin , LetsEncrypt, PHPswitcher


    I have to hide LetsEncrypt for the customers, because there is a malfunction somewhere atm.

    There are no indications of errors in the debugger and I can not find anything in the logs, maybe Fail2ban, UFW or IP_set blocked the IP addresses of Letsencrypt servers or something else is broken.

    Is there's a way to hide the Letsencrypt menu entry for the customers, so they are no longer able to create new certs? (cronjobs for cert renewals should stay).

    An incorrectly created certificate crashes my PHPswitcher (PHP Versions) at least 2 times a day and I have to delete and recreate it and all related customers settings. again & again ...

    For now I see no other option so I need to hide the menu entry currently I can not perform a reinstallation or upgrade due to illness.

    if anyone finds such tunxxx.socks files in /tmp, this is caused by a thirdparty PrestaShop database bridge.php which is trying to setup virtual proxy connections to their local computer accounting software, which also leads to a high system load. In MC you can set display info/details to the right side and then the respective vuXXXX owner are shown..

    OK, I have found the owner and creator of these .sock files and blocked their websites ... there are about 4 prestashops installed .... but I guess the webuser should not be able to generate TUN.SOCK files inside /tmp and it looks like hacked or outdated PHP scripts causing this - right?

    - IMSCP: 1.5.1

    - Distribution: Debian GNU/Linux 8.11 (jessie)

    - Proftpd

    - PHP FCGID

    - MariaDB 10.0

    - Courier

    - Roundcube

    - Web2FTP

    - Plugins:

    PMA Captcha, RoundCubePlugins, SpamAssassin , LetsEncrypt, WHMCS



    I found strange .sock files in / tmp (=tun7262.sock) with a filesize of 0, is this something I have to be worry about?

    How can I determine from which user these were generated?


    I know this installation is a little older will be migrated to a new server ASAP


    Screenshot-20190217-091911.png

    After the New Year party and a 3-day recovery period I was able to install (migrate) everything successfully in a VM!

    This time I have cleared the web_software tables and it worked without any problems ...

    Code
    1. truncate table web_software
    2. truncate table web_software_etc

    I have created a few own software packages for the imscp installer, could it be that the previous upgrade error was related to it?

    Code
    1. iMSCP::Stepper::_callback: DBD::mysql::db selectrow_hashref failed: No database selected at /usr/local/src/imscp-1.5.3-2018120800/engine/PerlLib/iMSCP/Dialog/InputValidation.pm line 487

    at the weekend I will try again the upgrade with the real system.