Closed TuringTux closed 1 year ago
It seems like the lines
- POSTGRESQL_USER=${POSTGRES_USER}
- POSTGRESQL_PASS=${POSTGRES_PASS}
need to be added to the environment
section of zammad-init
, zammad-scheduler
, zammad-railsserver
, zammad-websocket
. Update 2023-08-02: For the currently latest version of zammad, they also need to be added to zammad-nginx
why not use the ENV file.
Sorry, I do not understand what you mean.
I am in fact using the .env
file (as I always have been), which contains the variables POSTGRES_USER
and POSTGRES_PASS
just like in the example given in this repository.
However, it appears as those variables are not passed to the containers zammad-init
, zammad-scheduler
, zammad-railsserver
and zammad-websocket
anymore (where they are all needed), due to a missing declaration int the docker-compose.yml
file.
I am trying to deploy the Docker version in our server, and I'm having issues setting things up. Below you'll find logs for the part that fails.
@AnXh3L0 Without having looked to much into your problem, these were the changes I needed to make to docker-compose.yml
to get my instance up again:
diff --git a/docker-compose.yml b/docker-compose.yml
index b1640d0..8d207b5 100644
--- a/docker-compose.yml
+++ b/docker-compose.yml
@@ -35,6 +35,8 @@ services:
- zammad-postgresql
environment:
- MEMCACHE_SERVERS=${MEMCACHE_SERVERS}
+ - POSTGRESQL_USER=${POSTGRES_USER}
+ - POSTGRESQL_PASS=${POSTGRES_PASS}
- REDIS_URL=${REDIS_URL}
image: ${IMAGE_REPO}:${VERSION}
restart: on-failure
@@ -78,6 +80,8 @@ services:
- zammad-redis
environment:
- MEMCACHE_SERVERS=${MEMCACHE_SERVERS}
+ - POSTGRESQL_USER=${POSTGRES_USER}
+ - POSTGRESQL_PASS=${POSTGRES_PASS}
- REDIS_URL=${REDIS_URL}
image: ${IMAGE_REPO}:${VERSION}
restart: ${RESTART}
@@ -99,6 +103,8 @@ services:
environment:
- MEMCACHE_SERVERS=${MEMCACHE_SERVERS}
- REDIS_URL=${REDIS_URL}
+ - POSTGRESQL_USER=${POSTGRES_USER}
+ - POSTGRESQL_PASS=${POSTGRES_PASS}
image: ${IMAGE_REPO}:${VERSION}
restart: ${RESTART}
volumes:
@@ -112,6 +118,8 @@ services:
- zammad-redis
environment:
- MEMCACHE_SERVERS=${MEMCACHE_SERVERS}
+ - POSTGRESQL_USER=${POSTGRES_USER}
+ - POSTGRESQL_PASS=${POSTGRES_PASS}
- REDIS_URL=${REDIS_URL}
image: ${IMAGE_REPO}:${VERSION}
restart: ${RESTART}
@@ -120,13 +128,16 @@ services:
volumes:
elasticsearch-data:
- driver: local
+ external: true
+ name: zammad_elasticsearch-data
postgresql-data:
- driver: local
+ external: true
+ name: zammad_postgresql-data
redis-data:
driver: local
zammad-backup:
- driver: local
+ external: true
+ name: zammad_zammad-backup
zammad-config-nginx:
driver: local
zammad-var:
I needed to fix two unrelated problems:
POSTGRESQL_USER
and POSTGRESQL_PASS
variables (make sure they are also specified in the .env
) files for the services@TuringTux Thanks, I managed to get the instance to boot and load. Now I am trying to reverse proxy it because I use Nginx, but I can't seem to access the instance via https (it only responds on domain:8080) and I get CSRF error (I did try the solution from the docs regarding that, but no luck till now).
Hi there! Recently, there was a small change that could solve the reported behavior here: https://github.com/zammad/zammad/commit/6dc6b0fae5089a7a81b1a5f107f972b9ca279247
@AnXh3L0 CSRF sounds like you do not pass X-Forwarded-Proto headers to the nginx container,
Should be fixed by #366
Description
After upgrading to Zammad 6.x using the Docker configuration from this repository, the instance won't start up anymore, because the init container fails to connect to postgres.
We have specified a custom
POSTGRES_PASS
in.env
, which previously worked. I noticed that in the new release, thezammad-init
container doesn't have thePOSTGRESQL_USER
andPOSTGRESQL_PASS
variables set. Do I have to provide the password to the init container differently?Infos
Expected behavior
docker compose up
, Zammad starts up.Actual behavior
Zammad does not start up, but the init container does instead fail with the message:
Steps to reproduce
zammad
zammad-docker-compose