Hello,
I have a problem using your wordpress backup.
When running the backup script it's stopping before the end, the db archive is not created :

root@myhost# docker exec -it backup-container bash
root@f586c32ad7e9:/# bash backup

The log is ending on :

...
-rwxrwxr-x www-data/www-data      238 2019-05-21 13:26 ./wp-admin/ms-options.php
-rwxrwxr-x www-data/www-data     6320 2019-05-21 13:26 ./wp-admin/plugin-install.php
-rwxrwxr-x www-data/www-data      369 2019-05-21 13:26 ./wp-blog-header.php

When running throw the Makefile that is on my host the script runs well till the end :
(extract of my Makefile)

backup:
	docker exec backup-container backup

Any idea ?
Thanks for your work !

At the moment the mysql password is read from an env variable MYSQL_ENV_MYSQL_PASSWORD

I'm not using the MYSQL_ENV_MYSQL_PASSWORD or MYSQL_ENV_MYSQL_ROOT_PASSWORD env vars for setting passwords, instead I am using the alternative environment variables that end _FILE and poiint to a locally mounted secret file containing the password instead for better security.

This is described here (excerpt below): https://hub.docker.com/_/mysql/

Docker Secrets
As an alternative to passing sensitive information via environment variables, _FILE may be appended to the previously listed environment variables, causing the initialization script to load the values for those variables from files present in the container. In particular, this can be used to load passwords from Docker secrets stored in /run/secrets/<secret_name> files. For example:

$ docker run --name some-mysql -e MYSQL_ROOT_PASSWORD_FILE=/run/secrets/mysql-root -d mysql:tag
Currently, this is only supported for MYSQL_ROOT_PASSWORD, MYSQL_ROOT_HOST, MYSQL_DATABASE, MYSQL_USER, and MYSQL_PASSWORD.

Would it be possible to add support for checking for these _FILE environment variables and if present, obtaining the the password from the file they point to instead - as this will then allow me to continue to use secrets, which I'd quite like to continue doing.

How about using wp-cli to configure the wordpress??? The user could implement his own "wp-cli.yml" to uninstall, define hostname and etc.

Especially in case of import external sites for dev using docker.

How about it?

I used the docker image to my wordpress stack.
When I restore I see:

➜  wordtest docker exec wordtest_backup_1 restore 20181004                      
deleting files from /var/www/html/                                              
restoring files from /backups/backup_20181004.tar.gz to /var/www/html           
restoring data from mysql dump file /backups/backup_20181004.sql.bz2            
mysql: [Warning] Using a password on the command line interface can be insecure.
Finished: SUCCESS                                                               

As far as I know, that's the wrong approach for backup/restore and the whole disaster recovery.

• First you need to restore, make sure there is no error.
• Then delete the current data.
• Then move the restored data to the working directory.
  Not a big change and still simple.

Of course there is also other approach, but that's too mach for simple backup/restore:

• Restore the files to new dir (with data suffix for example)
• Rename original directory with .bak suffix
• Make the workdir a symlink to restored dir

You need to:
RUN apt-get update &&
apt-get install mysql-client cron -y &&
apt-get clean &&
rm -rf /var/lib/apt/lists/* &&
mkdir /backups

note that the official Docker wordpress and Mysql containers separate out their data directory (/var/www/html and /var/lib/mysql, I think). via the VOLUME command. Docker inspect the image to see where docker is saving your files.

So when you do a backup into a container, you're really backing up into the VOLUMEs that were created on your host, and not the actual containers.

Since the mysql update from 5.7.30 to 5.7.31 the backup script throws the following error, when executing the mysqldump statement:

mysqldump: Error: 'Access denied; you need (at least one of) the PROCESS privilege(s) for this operation' when trying to dump tablespaces

Also see the official release notes and this blog post.
As denoted here I proposa adding the "--no-tablespaces" option.

I am having problems using this in conjuction with jwilder/nginx-proxy ..

What I am attempting to do is a site restore from a bitnami wordpress install.

First I have tested that this works just fine out of the box with SSL using vanilla wordpress from Docker.

The issue arises when I attempt to restore my Wordpress from a Bitnami stack wordpress.

The install appears fine, until I enable SSL on Nginx-proxy. Then, it goes into a redirect loop.

If I re-copy the vanilla wp-config.php into the migrated image, I get redirected to https://127.0.0.1/wordpress/wp-admin/.

I have tried tinkering with wp-config.php with no real luck - although I am unsure how to enable the PHP logs on the docker vanilla image.

On my server I have only limited space and thus I'd like to have an alternative cleanup solution to the 'older than x days' solution. The alternative should be tiered and keep backups based on different periods. Before creating a pull request I'd like to have some feedback on my suggested implementation. Maybe there is a better one from point of usability. What I had in mind:

A new cli argument like '-CLEANUP_SCHEDULE=1.1.1.4.4.30.30'

The resulting backups plan would look like:

x = latest backup
- = deleted
Day 01 x
Day 02 x 1+4+30
Day 03 x      1 1+4+30
Day 04 x      1      1 1+4+30
Day 05 x      1      1      1   4+30
Day 06 x    1+4      1      1      -   4+30
Day 07 x      1    1+4      1      -      -   4+30
Day 08 x      1      1    1+4      -      -      -   4+30
Day 09 x      1      1      1      4      -      -      -   4+30
Day 10 x    1+4      1      1      -      4      -      -      -   4+30
Day 11 x      1    1+4      1      -      -      4      -      -      -   4+30
Day 12 x      1      1    1+4      -      -      -      4      -      -      -   4+30
Day 13 x      1      1      1      4      -      -      -      4      -      -      -    30
Day 14 x    1+4      1      1      -      4      -      -      -      -      -      -      -    30

The code would look like following:

# read periods from cli argument
var periods = [{distance: 1, count: 1}, {distance: 4, count: 2}, {distance: 30, count: 2}]
# read backup dates from file system
var backups = [19990101, 19990102, 19990103, 19990104, ...]

# cache which backups to keep and which to remove
for each backup:
  flag_for_deletion(backup)

# find out which backups to keep
for each period:
  for i from 1 to period.count:
    # find the best suited backup, i.e. the oldest backup that is not older than i * period.distance
    var best = backups.filter(backup is not older than i * period.distance).find(is oldest)
    unflag_for_deletion(best)

# remove unneeded backups
for each backup:
  if flagged_for_deletion(backup)
    delete backup

The tiered backup schedule is borrowed from https://hub.docker.com/r/prodrigestivill/postgres-backup-local. The latter uses the cli options BACKUP_KEEP_DAYS, BACKUP_KEEP_WEEKS, BACKUP_KEEP_MONTHS. But that control creates some duplicate backups which is not very convenient for my very limited storage situation.

Whenever I try to restore from backup I get this error

deleting files from /var/www/html/
rm: cannot remove '/var/www/html/wp-content': Device or resource busy```

I have tried stopping the Wordpress container and have gotten nothing. Any ideas why this would still be in use?

Can we make that Enviornment Variable?

Hi everyone.
I followed the tutorial and it worked. The problem is, I'm running and testing on Centos7 which is installed on Amazon EC2. After reboot instance and log in, i was not be able to restore back up.
It said;
Error response from daemon: Container wordpress-backup is not running
How should i do? If i run the wordpress-backup images, what would be the next step???

Hello,

This docker image is nice, a missing feature is that daily backups are never deleted: so this will just grow forever.

Did you happen to implement a solution to this, like a separate script to do some cleanup?

Alternatively, maybe an even cooler solution to this would be to implement incremental backup with tar. https://www.gnu.org/software/tar/manual/html_section/tar_40.html Just saying!

Thanks again for the great image 👍

Hi there,

Thx for your work :)

I try to make it works in a docker-stack.yml, but my containers names in my stack are partialy randomly created :

winter_wp.1.pexv7h9ifzr582og0jmdjq8ns

The solution, to me, is to find the way to find the variable part of the container name.

Can you help me to use your image in a swarm mode ?

thx :)

Getting this error... where do I need to start looking..?

Hi,
thanks for your handy container!
I encountered one issue.

The environment variables needed e.g. for restoring do not get set in my setup (pretty much like the readme, but mariadb instead of mysql)

If I manually set (in my docker-compose.yml)

MYSQL_ENV_MYSQL_USER: <WP-USER>
MYSQL_ENV_MYSQL_PASSWORD: <USER-PW>
MYSQL_ENV_MYSQL_DATABASE: <WP-DB>
MYSQL_PORT_3306_TCP_PORT: 3306

everything works fine.

I could not find where these environments should be generated from, so I can only guess this is some docker magic that I'm no yet aware of.

Any hint?

Hi,

First of all, thank you for your project.
I created a simple project where I customize a .env file and with docker-compose I create a WordPress website running on docker using MySQL, WordPress, Nginx-proxy (as a reverse proxy to access my websites on 80/443 port).
https://github.com/sasadangelo/webagency

Now I want to add a backup container. However, my understanding is that the backup is saved on a local volume I attach to the container. I am not an expert, but it would be helpful to have a system that saves these backups on S3 Object Storage, Dropbox, or other 3rdparty services.

I noticed there are several projects on the web, however, most of them simply sync the backup volumes with S3, for example. It would be better to have a container that keeps max N file in local (for example 5) and the rest on Object Storage. Moreover, it should have a sort of retention policy to say: keep only the latest N backup on S3.

Do you know if there is a project for that combined with your project meet these requirements?