[SOLVED]Dovecot issues on new ispconfig installation on vps Centos7

Zague · Dec 29, 2019

Hello guys,
I am following the perfect server guide to install ispconfig (https://www.howtoforge.com/tutorial...php-pureftpd-postfix-dovecot-and-ispconfig/2/). In the step to install dovecot after installation I tried to start the daemon and got this error:

Dec 28 23:10:43 mx1 systemd[1]: Starting Dovecot IMAP/POP3 email server...
Dec 28 23:10:43 mx1 systemd[1]: Can't open PID file /var/run/dovecot/master.pid (yet?) after start: Too many levels of symbolic links
Dec 28 23:10:43 mx1 dovecot[14112]: master: Dovecot v2.2.36 (1f10bfa63) starting up for imap, pop3, lmtp (core dumps disabled)
Dec 28 23:12:13 mx1 systemd[1]: dovecot.service start operation timed out. Terminating.
Dec 28 23:12:13 mx1 dovecot[14112]: master: Warning: Killed with signal 15 (by pid=1 uid=0 code=kill)
Dec 28 23:12:13 mx1 systemd[1]: Failed to start Dovecot IMAP/POP3 email server.
Dec 28 23:12:13 mx1 systemd[1]: Unit dovecot.service entered failed state.
Dec 28 23:12:13 mx1 systemd[1]: dovecot.service failed.
Click to expand...

The weird is if I start manually the dovecot with dovecot -c /etc/dovecot/dovecot.conf command, the daemon runs ok without issues, also
dovecot -n didn't give any errors or messages
dovecot -n

# 2.2.36 (1f10bfa63): /etc/dovecot/dovecot.conf
# Pigeonhole version 0.4.24 (124e06aa)
# OS: Linux 2.6.32-042stab127.2 x86_64 CentOS Linux release 7.7.1908 (Core)
first_valid_uid = 1000
mail_debug = yes
managesieve_notify_capability = mailto
managesieve_sieve_capability = fileinto reject envelope encoded-character vacation subaddress comparator-i;ascii-numeric relational regex imap4flags copy include variables body enotify environment mailbox date index ihave duplicate mime foreverypart extracttext
mbox_write_locks = fcntl
namespace inbox {
inbox = yes
location =
mailbox Drafts {
special_use = \Drafts
}
mailbox Junk {
special_use = \Junk
}
mailbox Sent {
special_use = \Sent
}
mailbox "Sent Messages" {
special_use = \Sent
}
mailbox Trash {
special_use = \Trash
}
prefix =
}
passdb {
driver = pam
}
plugin {
sieve = file:~/sieve;active=~/.dovecot.sieve
}
ssl = required
ssl_cert = </etc/pki/dovecot/certs/dovecot.pem
ssl_key = # hidden, use -P to show it
userdb {
driver = passwd
}
Click to expand...

Any Ideas?
Thank you

till · Dec 29, 2019

Follow the guide until the end and then test when you finished the installation, the dovecot config will get replaced anyway during ISPConfig installation.

Zague · Dec 29, 2019

Till,
I have completed the installation the issue is with 3 daemons dovecot,amavisd, pure-ftpd

Configuring DBServer
Installing ISPConfig crontab
[ERROR] Cron not found
Detect IP addresses
Restarting services ...
Job for amavisd.service failed because a timeout was exceeded. See "systemctl status amavisd.service" and "journalctl -xe" for details.
Job for dovecot.service failed because a timeout was exceeded. See "systemctl status dovecot.service" and "journalctl -xe" for details.
Job for pure-ftpd.service failed because a timeout was exceeded. See "systemctl status pure-ftpd.service" and "journalctl -xe" for details.
Installation completed.
Click to expand...

I did some research and the problem can be the kernel (Linux 2.6.32-042stab127.2 x86_64) but my vps provider does not allow kernel upgrade.
The message file reported
Dec 29 15:07:49 mx1 systemd: Can't open PID file /var/run/pure-ftpd.pid (yet?) after start: Too many levels of symbolic links
Dec 29 15:07:49 mx1 systemd: Can't open PID file /var/run/dovecot/master.pid (yet?) after start: Too many levels of symbolic links
Can't open PID file /var/run/amavisd/amavisd.pid (yet?) after start: Too many levels of symbolic links

Any Ideas

Jesse Norell · Dec 30, 2019

Zague said: ↑

Too many levels of symbolic links
Click to expand...

Sounds like you probably have symbolic links that point in a loop, check each level and you should find the problem.
Code:
ls -ld /var
ls -ld /var/run
ls -ld /var/run/amavis
ls -ld /var/run/amavis/amavis.pid
One or more of those will be a symbolic link, so also check the paths such links point to.

Zague · Dec 31, 2019

Till/Jesse
Thank you for yours answers, I have spend a lot of time trying to fix this, This is not my first installation so all aimed a problem with the image or the vps or kernel or something so I took the decision to switch to Ubuntu and everything was installed correct. In regular conditions I keep doing research and troubleshooting of the issue but now I need save time.
The last try that I did was update the systemd package and fix the problem of symlink but mariadb service was not able to start. That's why I switched the OS.
Thank you very much
Regards

Zague · Feb 6, 2020

Hello Guys,
Sorry to reopen this thread but today this happens again when everything was working since December, my OS is Ubuntu 16.4, today did a systemctl restart dovecot and the daemon didn't start with the same issue

root@mx1:/var/run/dovecot# systemctl status dovecot.service
● dovecot.service - Dovecot IMAP/POP3 email server
Loaded: loaded (/lib/systemd/system/dovecot.service; enabled; vendor preset: enabled)
Active: activating (start) since Thu 2020-02-06 09:10:29 EST; 5s ago
Docs: man:dovecot(1)
http://wiki2.dovecot.org/
Process: 18070 ExecStop=/usr/bin/doveadm stop (code=exited, status=0/SUCCESS)
Process: 20301 ExecStart=/usr/sbin/dovecot (code=exited, status=0/SUCCESS)
Main PID: 28533 (code=exited, status=0/SUCCESS)
CGroup: /system.slice/dovecot.service
├─20302 /usr/sbin/dovecot
├─20303 dovecot/anvil
├─20304 dovecot/log
└─20306 dovecot/config

Feb 06 09:10:29 mx1 systemd[1]: Starting Dovecot IMAP/POP3 email server...
Feb 06 09:10:29 mx1 dovecot[20302]: master: Dovecot v2.2.22 (fe789d2) starting up for imap, pop3 (core dumps disabled)
Feb 06 09:10:29 mx1 systemd[1]: dovecot.service: Can't open PID file /var/run/dovecot/master.pid (yet?) after start: Too many levels of symbolic links
Click to expand...

root@mx1:/var/run/dovecot# ls -ld /var
drwxr-xr-x 14 root root 4096 Dec 31 00:01 /var
root@mx1:/var/run/dovecot# ls -ld /var/run
lrwxrwxrwx 1 root root 4 Dec 29 22:33 /var/run -> /run
root@mx1:/var/run/dovecot# ls -ld /var/run/dovecot
drwxr-xr-x 5 root root 720 Feb 6 09:11 /var/run/dovecot
ls -ld /var/run/dovecot/master.pid
root@mx1:/var/run/dovecot#-rw------- 1 root root 6 Feb 6 09:04 /var/run/dovecot/master.pid
Click to expand...

I tried changing from PIDFILE=${PIDBASE:-/var/run/dovecot}/master.pid to PIDFILE=${PIDBASE:-/run/dovecot}/master.pid in /etc/init.d/dovecot, didn't work.
So any Ideas?
Thank you so much for your time

Jesse Norell · Feb 6, 2020

Curious, that does not show recursive symbolic links. What do you get for
Code:
ls -la /run
mount | grep run
systemctl --state=failed
Zague said: ↑

I tried changing from PIDFILE=${PIDBASE:-/var/run/dovecot}/master.pid to PIDFILE=${PIDBASE:-/run/dovecot}/master.pid in /etc/init.d/dovecot, didn't work.
Click to expand...

That init.d file won't be in use with systemd.

Zague · Feb 7, 2020

Thank you Jesse for your answer
here the output

root@mx1:~# ls -la /run
total 36
drwxr-xr-x 27 root root 840 Feb 7 10:27 .
drwxr-xr-x 23 root root 4096 Feb 6 09:20 ..
-rw------- 1 root root 0 Feb 6 09:20 agetty.reload
drwxr-xr-x 2 amavis amavis 60 Feb 7 10:22 amavis
drwxr-xr-x 2 root root 80 Feb 7 02:00 apache2
drwxr-xr-x 2 clamav root 60 Feb 6 09:21 clamav
-rw-r--r-- 1 root root 4 Feb 6 09:20 crond.pid
---------- 1 root root 0 Feb 6 09:20 crond.reboot
drwxr-xr-x 5 root root 700 Feb 7 11:18 dovecot
drwxr-xr-x 2 root root 40 Feb 7 11:52 fail2ban
lrwxrwxrwx 1 root root 25 Feb 6 09:20 initctl -> /run/systemd/initctl/fifo
drwxrwxrwt 6 root root 140 Feb 6 09:20 lock
drwxr-xr-x 3 root root 60 Feb 6 09:20 log
drwxr-xr-x 2 list list 60 Feb 6 09:20 mailman
-rw-r--r-- 1 root root 213 Feb 7 10:12 motd.dynamic
drwxr-xr-x 2 root root 60 Feb 6 09:20 mount
drwxr-xr-x 2 mysql root 80 Feb 6 09:20 mysqld
drwxrwxr-x 2 root root 40 Feb 6 09:20 named
drwxr-xr-x 2 root root 160 Feb 6 09:20 network
-rw-r--r-- 1 root root 3 Feb 6 09:21 ntpd.pid
drwxr-xr-x 2 opendmarc opendmarc 80 Feb 6 09:20 opendmarc
drwxr-xr-x 2 root root 80 Feb 6 09:20 php
-rw-r--r-- 1 postgrey postgrey 4 Feb 6 09:20 postgrey.pid
drwx------ 2 root root 60 Feb 7 12:05 pure-ftpd
drwxr-xr-x 2 root root 40 Feb 6 09:20 rpcbind
-r--r--r-- 1 root root 0 Feb 6 09:20 rpcbind.lock
srw-rw-rw- 1 root root 0 Feb 6 09:20 rpcbind.sock
-rw-r--r-- 1 root root 3 Feb 6 09:20 rsyslogd.pid
drwxr-xr-x 2 root root 40 Feb 6 09:20 samba
drwx--x--- 2 root sasl 140 Feb 6 09:20 saslauthd
drwxr-xr-x 2 root root 40 Feb 6 09:20 sendsigs.omit.d
drwxrwxrwt 2 root root 40 Feb 6 09:20 shm
drwxr-xr-x 2 root root 40 Feb 7 10:06 sshd
-rw-r--r-- 1 root root 6 Feb 7 10:07 sshd.pid
drwxr-xr-x 3 root root 60 Feb 7 10:07 sudo
drwxr-xr-x 16 root root 400 Feb 7 11:52 systemd
drwxr-xr-x 4 root root 100 Feb 6 09:20 udev
-rw-r----- 1 root root 0 Feb 7 06:45 unattended-upgrades.lock
drwxr-xr-x 2 root root 40 Feb 6 09:20 user
-rw-rw-r-- 1 root root 2304 Feb 7 10:12 utmp
-rw-r--r-- 1 root root 4 Feb 6 09:20 xinetd.pid
-rw------- 1 root root 0 Feb 6 09:20 xtables.lock
root@mx1:~#
Click to expand...

root@mx1:~# mount | grep run
tmpfs on /run type tmpfs (rw,nosuid,nodev,mode=755)
tmpfs on /run/lock type tmpfs (rw,nosuid,nodev,noexec,relatime,size=5120k)
none on /run/shm type tmpfs (rw,relatime)
Click to expand...

root@mx1:~# systemctl --state=failed
UNIT LOAD ACTIVE SUB DESCRIPTION
● dovecot.service loaded failed failed Dovecot IMAP/POP3 email server
● systemd-tmpfiles-setup.service loaded failed failed Create Volatile Files and Directories
● vz.service loaded failed failed LSB: OpenVZ startup script
● vzeventd.service loaded failed failed LSB: start and stop vzeventd

LOAD = Reflects whether the unit definition was properly loaded.
ACTIVE = The high-level unit activation state, i.e. generalization of SUB.
SUB = The low-level unit activation state, values depend on unit type.

4 loaded units listed. Pass --all to see loaded but inactive units, too.
To show all installed unit files use 'systemctl list-unit-files'.
root@mx1:~#
Click to expand...

I found the below link, it didn't fit exactly with my cases but I tried to create manually the links but it didn't work
https://github.com/systemd/systemd/issues/3010
Thank you for your time
Regards

Zague · Feb 7, 2020

I think the issue is related with systemd because I ran manually daemons and they start ok
dovecot -c /etc/dovecot/dovecot.conf

root@mx1:/usr/lib/tmpfiles.d# ps -feal|grep dovecot
1 S root 1442 1 0 80 0 - 4527 ep_pol 13:24 ? 00:00:00 /usr/sbin/dovecot -c /etc/dovecot/dovecot.conf
4 S dovecot 1443 1442 0 80 0 - 2366 ep_pol 13:24 ? 00:00:00 dovecot/anvil
4 S root 1444 1442 0 80 0 - 2399 ep_pol 13:24 ? 00:00:00 dovecot/log
4 S root 3882 1442 0 80 0 - 11659 ep_pol 14:10 ? 00:00:00 dovecot/auth
4 S root 3885 1442 0 80 0 - 2365 ep_pol 14:10 ? 00:00:00 dovecot/ssl-params
0 S root 3943 1147 0 80 0 - 2806 pipe_w 14:12 pts/0 00:00:00 grep --color=auto dovecot
Click to expand...

The same situation is happening with fail2ban but running manually it's working ok
fail2ban-client -x start

root@mx1:/usr/lib/tmpfiles.d# ps -feal|grep fail2ban
5 S root 1707 1 1 80 0 - 412944 poll_s 13:26 ? 00:00:49 /usr/bin/python3 /usr/bin/fail2ban-server -s /var/run/fail2ban/fail2ban.sock -p /var/run/fail2ban/fail2ban.pid -x -b
0 S root 3983 1147 0 80 0 - 2806 pipe_w 14:14 pts/0 00:00:00 grep --color=auto fail2ban
Click to expand...

I am stuck and I don't know what else to do
Any ideas?

till · Feb 7, 2020

What kind of VPN is it and how much RAM do you have?

Zague · Feb 7, 2020

Till
VPS:OpenVZ
Memory 3Gb/Used 70% 2.11
HDD 100GB/Used 2.78GB
Bandwidth: 3TB /Used 410.29 MB
Template:Ubuntu 16.04 X86 64
--------------------------------
In the below link there is a workaround making downgrade to systemd
https://bugs.launchpad.net/ubuntu/+source/systemd/+bug/1804847

wget http://launchpadlibrarian.net/395488411/systemd_229-4ubuntu21.6_amd64.deb | wget http://launchpadlibrarian.net/395488401/libpam-systemd_229-4ubuntu21.6_amd64.deb | wget http://launchpadlibrarian.net/395488403/libsystemd0_229-4ubuntu21.6_amd64.deb
dpkg -i libpam-systemd_229-4ubuntu21.6_amd64.deb systemd_229-4ubuntu21.6_amd64.deb libsystemd0_229-4ubuntu21.6_amd64.deb
apt-mark hold systemd libsystemd0 libpam-systemd
Click to expand...

But I don't know how this procedure can affect ispconfig or others packages.

Zague · Feb 7, 2020

Guys,
I was able to fix the issue according with all information that I found I reviewed the file "/etc/systemd/system/multi-user.target.wants/fail2ban.service" and change the lines
Code:
from
PIDFile=/var/run/fail2ban/fail2ban.pid
to
PIDFile=/run/fail2ban/fail2ban.pid
I did the same with dovecot nano /etc/systemd/system/multi-user.target.wants/dovecot.service
Code:
from
PIDFile=/var/run/dovecot/master.pid
to
PIDFile=/run/dovecot/master.pid
I think that this workaround also could be applied in centOS in last December when this issue presented at installation stage, but in that time I didn't have enough time to make research.
Thank you guys for the efforts and time dedicated to this thread, we can mark this thread as solved

till · Feb 8, 2020

Thank you for posting the solution!

Log in or Sign up

[SOLVED]Dovecot issues on new ispconfig installation on vps Centos7

Zague Member

till Super Moderator Staff Member ISPConfig Developer

Zague Member

Jesse Norell Well-Known Member Staff Member Howtoforge Staff

Zague Member

Zague Member

Jesse Norell Well-Known Member Staff Member Howtoforge Staff

Zague Member

Zague Member

till Super Moderator Staff Member ISPConfig Developer

Zague Member

Zague Member

till Super Moderator Staff Member ISPConfig Developer

Share This Page

Log in or Sign up

[SOLVED]Dovecot issues on new ispconfig installation on vps Centos7

Zague Member

till Super Moderator Staff Member ISPConfig Developer

Zague Member

Jesse Norell Well-Known Member Staff Member Howtoforge Staff

Zague Member

Zague Member

Jesse Norell Well-Known Member Staff Member Howtoforge Staff

Zague Member

Zague Member

till Super Moderator Staff Member ISPConfig Developer

Zague Member

Zague Member

till Super Moderator Staff Member ISPConfig Developer

Share This Page

Useful Searches