Dovecot crashes and isn't restarted properly

Issue #110 closed
Former user created an issue

Hi,

For a little while now I've had an issue with Dovecot crashing and not being restarted.

I first noticed it when Roundcube gave me a "Error connecting to storage" when I tried to log in my mailbox. Today I finally decided to look into it a bit more so I entered the container with "docker exec -ti container /bin/bash" in order to look around.

First, here's the error message in /var/log/message :

root@c50704bd7d18:/# tail -5 /var/log/messages
Aug 20 07:43:41 c50704bd7d18 mail.crit dovecot: master: Fatal: Dovecot is already running with PID 538 (read from /var/run/dovecot/master.pid)
Aug 20 07:43:42 c50704bd7d18 mail.crit dovecot: master: Fatal: Dovecot is already running with PID 538 (read from /var/run/dovecot/master.pid)
Aug 20 07:43:43 c50704bd7d18 mail.crit dovecot: master: Fatal: Dovecot is already running with PID 538 (read from /var/run/dovecot/master.pid)
Aug 20 07:43:44 c50704bd7d18 mail.crit dovecot: master: Fatal: Dovecot is already running with PID 538 (read from /var/run/dovecot/master.pid)
Aug 20 07:43:45 c50704bd7d18 mail.crit dovecot: master: Fatal: Dovecot is already running with PID 538 (read from /var/run/dovecot/master.pid)

There's no dovecot process and the process with a pid of 538 is an s6-log process :

root@c50704bd7d18:/#  ps -aux | grep -v grep | grep dovecot
root       529  0.0  0.0    200     4 ?        S+   Aug19   0:00 s6-supervise dovecot/log
root       530  0.0  0.0    200    52 ?        S+   Aug19   0:29 s6-supervise dovecot
nobody     557  0.0  0.0    260     4 ?        Ss   Aug19   0:00 s6-log -bp -- n20 s1000000 T /var/log/s6/dovecot
root@c50704bd7d18:/# ps -aux | grep -v grep | grep 538
nobody     538  0.0  0.0    260     4 ?        Ss   Aug19   0:00 s6-log -bp -- n20 s1000000 T /var/log/s6/clamd

I'm starting the container using docker-compose with the following compose file :

mailserver:
  image: analogic/poste.io
  ports:
    - "25:25"
    - "127.0.0.1:1080:80"
    - "127.0.0.1:1443:443"
    - "110:110"
    - "143:143"
    - "465:465"
    - "587:587"
    - "993:993"
    - "995:995"
  volumes:
    - /etc/localtime:/etc/localtime:ro
    - /opt/container-volumes/poste.io:/data
  restart: always
  tty: true

Running the poste.io image from August 9th : [root@ns3453436 ~]# docker images REPOSITORY TAG IMAGE ID CREATED SIZE docker.io/analogic/poste.io latest f3a24f554fd0 11 days ago 476.3 MB

Comments (3)

  1. Trexmaster Trexmaster

    Forgot to log in when I created that issue, to add some other info recreating the container solves the issue (that is, until the next crash), so does restarting the container or the host.

  2. SH repo owner

    Do you have clue why Dovecot crashed? Do you have low memory setup? There should be log in .../log/syslog

  3. Log in to comment