Similar problem here, but I have more information to track this down (I use my letsencrypt certificates from the linuxserver image as you described in the GitHub wiki):
My error message:
[INFO] Let's encrypt live directory found
[INFO] Using /etc/letsencrypt/live/mail.MYDOMAIN.net folder
[INFO] Found DKIM key pair for domain MYDOMAIN.net - skip creation
[INFO] MariaDB hostname not found in /etc/hosts
[ERROR] Container IP not found with embedded DNS server... Abort !
I think the problem lies in "run.sh" in either the the variable assignment or the added entry to /etc/hosts:
DBHOST=${MYSQL_HOST:-mariadb}
# Check mariadb hostname
grep -q "${DBHOST}" /etc/hosts
if [ $? -ne 0 ]; then
echo "[INFO] MariaDB hostname not found in /etc/hosts"
IP=$(dig A ${DBHOST} +short)
if [ -n "$IP" ]; then
echo "[INFO] Container IP found, adding a new record in /etc/hosts"
echo "${IP} ${DBHOST}" >> /etc/hosts
else
echo "[ERROR] Container IP not found with embedded DNS server... Abort !"
exit 1
fi
else
echo "[INFO] MariaDB hostname found in /etc/hosts"
fi
Since i don't know how to get into this container on startup and don't run a new instance on startup I'm stuck a little.
Can you investigate?
Improvement Idea: Also the template is missing the option for DBPORT, as for me all db's are running in separate containers with different ports.
EDIT:
Ok, found some more infos:
1. you can get the container to start and don't stop right away if you set the network option to "none" and then "docker exec -it malfiurios-mailserver sh" to get inside
2. the error message in /etc/hosts is
; <<>> DiG 9.10.3-P4-Debian <<>> A <MY-MYSQL-IP> +short
;; global options: +cmd
;; connection timed out; no servers could be reached <MY-MYSQL-IP>
; <<>> DiG 9.10.3-P4-Debian <<>> A 127.0.0.2 +short
;; global options: +cmd
;; connection timed out; no servers could be reached 127.0.0.2
Since I read your readme all the way to the end (;-)) I found the hardware/mailserver GitHub and this issue there: https://github.com/hardware/mailserver/issues/150. Sadly the answer involves some config changes for docker-compose which we can't use on unraid.
The issue for your postfixadmin template here https://forums.lime-technology.com/topic/60111-support-malfurious-roundcube-postfixadmin/?do=findComment&comment=608953 seems to be the same, so maybe we can solve both.