Bootloop on version 2.3.4

Issue #912 new
Ulrich Van Den Hekke created an issue

After some weeks of usage, and a restart, poste.io bootloop.

The container is killed after some seconds of running.

If i look on the healthcheck i have :

Clamd is running
Rspamd is running
Cron is running
Dovecot is running
NGiNX is running
p0f is running
PHP-fpm is running
Redis is running
SMTP is not running

sometimes i have Submission is not running too.

In the log of haraka-smtp i have :

2021-12-15 23:45:44.338645500  [DEBUG] [-] [core] loading dhparams from dhparams.pem
2021-12-15 23:45:44.338667500  [CRIT] [-] [core] Error: error:0909006C:PEM routines:get_name:no start line
2021-12-15 23:45:44.338688500  [CRIT] [-] [core]     at Object.createSecureContext (_tls_common.js:125:19)
2021-12-15 23:45:44.338711500  [CRIT] [-] [core]     at /usr/lib/node_modules/Haraka/tls_socket.js:416:34
2021-12-15 23:45:44.338733500  [CRIT] [-] [core]     at Object.exports.ensureDhparams (/usr/lib/node_modules/Haraka/tls_socket.js:549:16)
2021-12-15 23:45:44.338756500  [CRIT] [-] [core]     at Object.exports.load_default_opts (/usr/lib/node_modules/Haraka/tls_socket.js:415:14)
2021-12-15 23:45:44.338777500  [CRIT] [-] [core]     at Object.exports.load_tls_ini (/usr/lib/node_modules/Haraka/tls_socket.js:290:14)
2021-12-15 23:45:44.338800500  [CRIT] [-] [core]     at Plugin.exports.register (/usr/lib/node_modules/Haraka/plugins/tls.js:11:16)
2021-12-15 23:45:44.338822500  [CRIT] [-] [core]     at Object.plugins._register_plugin (/usr/lib/node_modules/Haraka/plugins.js:404:12)
2021-12-15 23:45:44.338843500  [CRIT] [-] [core]     at Object.plugins.load_plugin (/usr/lib/node_modules/Haraka/plugins.js:378:17)
2021-12-15 23:45:44.338867500  [CRIT] [-] [core]     at /usr/lib/node_modules/Haraka/plugins.js:326:21
2021-12-15 23:45:44.338888500  [CRIT] [-] [core]     at Array.forEach (<anonymous>)
2021-12-15 23:45:44.339680500  [NOTICE] [-] [core] Shutting down
2021-12-15 23:45:45.059641500  loaded TLD files:

What can i do to resolve the problems.

Comments (1)

  1. Mattia

    i had the same problems when the container or host did not have the correct time.
    once set you have to delete the never queue from the directory before starting the container (because they have a previous or past sending date)

  2. Log in to comment