127 lines
4.7 KiB
Markdown
127 lines
4.7 KiB
Markdown
## Install mailcow
|
|
|
|
**WARNING**: Please use Ubuntu 16.04 instead of Debian 8 or [switch to the kernel 4.9 from jessie backports](https://packages.debian.org/jessie-backports/linux-image-amd64) because there is a bug (kernel panic) with the kernel 3.16 when running docker containers with healthchecks! Full details here: [github.com/docker/docker/issues/30402](https://github.com/docker/docker/issues/30402) and [forum.mailcow.email/t/solved-mailcow-docker-causes-kernel-panic-edit/448](https://forum.mailcow.email/t/solved-mailcow-docker-causes-kernel-panic-edit/448)
|
|
|
|
You need Docker and Docker Compose.
|
|
|
|
1\. Learn how to install [Docker](https://docs.docker.com/engine/installation/linux/) and [Docker Compose](https://docs.docker.com/compose/install/).
|
|
|
|
Quick installation for most operation systems:
|
|
|
|
- Docker
|
|
```
|
|
curl -sSL https://get.docker.com/ | sh
|
|
```
|
|
|
|
- Docker-Compose
|
|
```
|
|
curl -L https://github.com/docker/compose/releases/download/$(curl -Ls https://www.servercow.de/docker-compose/latest.php)/docker-compose-$(uname -s)-$(uname -m) > /usr/local/bin/docker-compose
|
|
chmod +x /usr/local/bin/docker-compose
|
|
```
|
|
|
|
Please use the latest Docker engine available and do not use the engine that ships with your distros repository.
|
|
|
|
2\. Clone the master branch of the repository
|
|
```
|
|
git clone https://github.com/andryyy/mailcow-dockerized && cd mailcow-dockerized
|
|
```
|
|
|
|
3\. Generate a configuration file. Use a FQDN (`host.domain.tld`) as hostname when asked.
|
|
```
|
|
./generate_config.sh
|
|
```
|
|
|
|
4\. Change configuration if you want or need to.
|
|
```
|
|
nano mailcow.conf
|
|
```
|
|
If you plan to use a reverse proxy, you can, for example, bind HTTPS to 127.0.0.1 on port 8443 and HTTP to 127.0.0.1 on port 8080.
|
|
|
|
You may need to stop an existing pre-installed MTA which blocks port 25/tcp. See [this chapter](https://andryyy.github.io/mailcow-dockerized/first_steps/#install-a-local-mta) to learn how to reconfigure Postfix to run besides mailcow after a successful installation.
|
|
|
|
5\. Pull the images and run the composer file. The paramter `-d` will start mailcow: dockerized detached:
|
|
```
|
|
docker-compose pull
|
|
docker-compose up -d
|
|
```
|
|
|
|
Done!
|
|
|
|
You can now access **https://${MAILCOW_HOSTNAME}** with the default credentials `admin` + password `moohoo`.
|
|
|
|
The database will be initialized right after a connection to MySQL can be established.
|
|
|
|
## Update mailcow
|
|
|
|
There is no update routine. You need to refresh your pulled repository clone and apply your local changes (if any). Actually there are many ways to merge local changes.
|
|
|
|
### Step 1, method 1
|
|
Stash all local changes, pull changes from the remote master branch and apply your stash on top of it. You will most likely see warnings about non-commited changes; you can ignore them:
|
|
|
|
```
|
|
# Stash local changes
|
|
git stash
|
|
# Re-pull master
|
|
git pull
|
|
# Apply stash and remove it
|
|
git stash pop
|
|
```
|
|
|
|
### Step 1, method 2
|
|
Fetch new data from GitHub, commit changes and merge remote repository:
|
|
|
|
```
|
|
# Get updates/changes
|
|
git fetch
|
|
# Add all changed files to local clone
|
|
git add -A
|
|
# Commit changes, ignore git complaining about username and mail address
|
|
git commit -m "Local config aat $(date)"
|
|
# Merge changes
|
|
git merge
|
|
```
|
|
|
|
If git complains about conflicts, solve them! Example:
|
|
```
|
|
CONFLICT (content): Merge conflict in data/web/index.php
|
|
```
|
|
|
|
Open `data/web/index.php`, solve the conflict, close the file and run `git add -A` + `git commit -m "Solved conflict"`.
|
|
|
|
### Step 1, method 3
|
|
|
|
Thanks to fabreg @ GitHub!
|
|
|
|
In case both methods do not work (for many reason like you're unable to fix the CONFLICTS or any other reasons) you can simply start all over again.
|
|
|
|
Keep in mind that all local changes _to configuration files_ will be lost. However, your volumes will not be removed.
|
|
|
|
- Copy mailcow.conf somewhere outside the mailcow-dockerized directory
|
|
- Stop and remove mailcow containers: `docker-compose down`
|
|
- Delete the directory or rename it
|
|
- Clone the remote repository again (`git clone https://github.com/andryyy/mailcow-dockerized && cd mailcow-dockerized`). **Pay attention** to this step - the folder must have the same name of the previous one!
|
|
- Copy back your previous `mailcow.conf` into the mailcow-dockerizd folder
|
|
|
|
If you forgot to stop Docker before deleting the cloned directoy, you can use the following commands:
|
|
```
|
|
docker stop $(docker ps -a -q)
|
|
docker rm $(docker ps -a -q)
|
|
```
|
|
|
|
### Step 2
|
|
|
|
Pull new images (if any) and recreate changed containers:
|
|
|
|
```
|
|
docker-compose pull
|
|
docker-compose up -d --remove-orphans
|
|
```
|
|
|
|
### Step 3
|
|
Clean-up dangling (unused) images and volumes:
|
|
|
|
```
|
|
docker rmi -f $(docker images -f "dangling=true" -q)
|
|
docker volume rm $(docker volume ls -qf dangling=true)
|
|
```
|