Upgrade to The username is already being used

Configuration is as follows:
OWNCLOUD_DOMAIN=yp.mapbar.com
OWNCLOUD_ADMIN_USERNAME=admin
OWNCLOUD_ADMIN_PASSWORD=yp111qqq,
OWNCLOUD_VERSION=10.0.8
OWNCLOUDPATH=/mapbar/data/owncloud/data/owncloud
docker run -d
–name owncloud
–link mariadb:db
–link redis:redis
-p 80:80
-p 443:443
-e OWNCLOUD_DOMAIN=${OWNCLOUD_DOMAIN}
-e OWNCLOUD_DB_TYPE=mysql
-e OWNCLOUD_DB_NAME=owncloud
-e OWNCLOUD_DB_USERNAME=owncloud
-e OWNCLOUD_DB_PASSWORD=owncloud
-e OWNCLOUD_DB_HOST=db
-e OWNCLOUD_ADMIN_USERNAME=${OWNCLOUD_ADMIN_USERNAME}
-e OWNCLOUD_ADMIN_PASSWORD=${OWNCLOUD_ADMIN_PASSWORD}
-e OWNCLOUD_REDIS_ENABLED=true
-e OWNCLOUD_REDIS_HOST=redis
-v $OWNCLOUDPATH:/mnt/data:z
-v /mapbar/data/owncloud/data/autoai-sslkey:/mnt/data/certs
owncloud/server:${OWNCLOUD_VERSION}

After upgrading from 10.0.8 to 10.0.10, start a new mirror and report an error
Restore to 10.0.8 and the startup still reports an error
Reconstruction can be resolved, but the configuration is lost

What error in particular?

This is the container’s error log:

Creating volume folders…
Creating hook folders…
Waiting for MySQL…
wait-for-it: waiting 180 seconds for db:3306
wait-for-it: db:3306 is available after 0 seconds
Waiting for Redis…
wait-for-it: waiting 60 seconds for redis:6379
wait-for-it: redis:6379 is available after 0 seconds
Removing config folder…
Linking config folder…
Writing config file…
Fixing base perms…
Fixing data perms…
Fixing hook perms…
Installing server database…
The username is already being used

Who can help me solve it

You have to delete the already existing db user owncloud.

1 Like

This way, the library is removed every time you upgrade or restart

image
The directory that maps out looks like this. Is that correct

It is not until now that I upgraded that I had a secondary problem. Every time I restart the image, there will be this problem. Is it the docker volume mount problem


The problem has been pointed out, but no solution has been found

2 Likes

I have the same problem. How I fix this? I don’t want remove user (and meaby reset database) every time I need to restart the docker instance…

Very bad practice if you have to really do this… Why the database initialization can’t just work “insert if not exists” like?

1 Like

I have the same problem. Does anyone have a solution? I don’t want to reset and reconfigure the whole database just to restart the container…