How to upgrade a Script Installation from 9.3.7 to 10?

[Gomez]

I've always had problems upgrading my script installation but I'm currently on version 9.3.7

I just tried to upgrade to the latest version using

sudo /var/www/espocrm/command.sh upgrade

and unsurprisingly nothing updated.

after sudo /var/www/espocrm/command.sh logs

I see

warning: LEGACY INSTALLATION METHOD DETECTED.
warning: Do not mount /var/www/html directly. Instead, mount the following directories separately:
warning: /var/www/html/custom
warning: /var/www/html/data
warning: /var/www/html/client/custom
warning: No further EspoCRM upgrades will be available.
warning: See https://docs.espocrm.com/administrat...-to-espocrm-10

When I check that page it's all about Migration when its "Installation with Docker" and there are options for local directory or docker volumes.

If I select "Installation by script" on the left there is nothing about migration to 10.

Sorry if I'm being stupid but how do I upgrade to 10 from a script installation?

[heint]

Greetings, Gomez

It seems like this inconvenience during the upgrade happened because you used the "installation by script" script before v.10 of EspoCRM had been released. Users who installed Espo by script after v.10 was released are not supposed to face this problem.

For your case it's recommended to follow the instruction from the log (migration), but I'm strongly emphasizing making a backup beforehand. All your files are located at /var/www/espocrm path.

[Gomez]

Thanks for the reply heint

Before I commit can I just check where to modify my docker-compose.yml correctly?

In the current file (see below) the volumes: section is under espocrm-nginx: section and contains the line

- ./data/espocrm:/var/www/html

Do I follow the local directory instructions and change this to :-

- ./data/espocrm/data:/var/www/espocrm/data
- ./data/espocrm/custom:/var/www/espocrm/custom
- ./data/espocrm/client/custom:/var/www/espocrm/cliet/custom

And the same under espocrm: amd espocrm-daemon: and espocrm-websocket: sections?

After I've done that do I then just run sudo /var/www/espocrm/command.sh upgrade again?

I see there this an optional step in the instructions to remove obsolete files which appears (amongst other things) to delete command.php, which is called from command,sh - should I not do this?

I also notice that image listed under the espocrm section is espocrm/espocrm:fpm but the instructions just show espocrm/espocrm is there any need to change that.

I'm sorry for all the questions but I really want to get this right before I attempt it.

current docker-compose.yml follows

# MODE: http
# VERSION: 2.5.1

services:

espocrm-db:
image: mariadb:latest
container_name: espocrm-db
command: --max-allowed-packet=64MB
restart: always
environment:
MARIADB_ROOT_PASSWORD: {root-passowrd}
MARIADB_DATABASE: espocrm
MARIADB_USER: {username}
MARIADB_PASSWORD: {dbpassword}
volumes:
- ./data/mariadb/data:/var/lib/mysql
networks:
- espocrm-network
healthcheck:
test: ["CMD", "healthcheck.sh", "--connect", "--innodb_initialized"]
interval: 20s
start_period: 10s
timeout: 10s
retries: 3

espocrm-nginx:
image: nginx:latest
container_name: espocrm-nginx
environment:
NGINX_HOST: 192.168.38.5
volumes:
- ./data/nginx/conf.d/:/etc/nginx/templates
- ./data/espocrm:/var/www/html
- ./data/nginx/espocrm.conf:/etc/nginx/espocrm.conf
- ./data/nginx/logs:/var/log/nginx
restart: always
depends_on:
- espocrm
- espocrm-websocket
ports:
- "8081:80"
networks:
- espocrm-network

espocrm:
image: espocrm/espocrm:fpm
container_name: espocrm
environment:
ESPOCRM_DATABASE_HOST: espocrm-db
ESPOCRM_DATABASE_USER: {user}
ESPOCRM_DATABASE_PASSWORD: {database-password}
ESPOCRM_ADMIN_USERNAME: {admin-user}
ESPOCRM_ADMIN_PASSWORD: {admin-password}
ESPOCRM_CONFIG_SITE_URL: "http://192.168.38.5"
restart: always
depends_on:
espocrm-db:
condition: service_healthy
volumes:
- ./data/espocrm:/var/www/html
networks:
- espocrm-network


espocrm-daemon:
image: espocrm/espocrm:fpm
container_name: espocrm-daemon
volumes:
- ./data/espocrm:/var/www/html
restart: always
depends_on:
- espocrm
entrypoint: docker-daemon.sh
networks:
- espocrm-network

espocrm-websocket:
container_name: espocrm-websocket
image: espocrm/espocrm:fpm
environment:
ESPOCRM_CONFIG_USE_WEB_SOCKET: "true"
ESPOCRM_CONFIG_WEB_SOCKET_ZERO_M_Q_SUBSCRIBER_DSN: "tcp://*:7777"
ESPOCRM_CONFIG_WEB_SOCKET_ZERO_M_Q_SUBMISSION_DSN: "tcp://espocrm-websocket:7777"
volumes:
- ./data/espocrm:/var/www/html
restart: always
depends_on:
- espocrm
entrypoint: docker-websocket.sh
ports:
- "8080:8080"
networks:
- espocrm-network

networks:
espocrm-network:
driver: bridge

[heint]

Gomez,

In the current file (see below) the volumes: section is under espocrm-nginx: section and contains the line

- ./data/espocrm:/var/www/html

Do I follow the local directory instructions and change this to :-

- ./data/espocrm/data:/var/www/espocrm/data
- ./data/espocrm/custom:/var/www/espocrm/custom
- ./data/espocrm/client/custom:/var/www/espocrm/cliet/custom

And the same under espocrm: amd espocrm-daemon: and espocrm-websocket: sections?

After I've done that do I then just run sudo /var/www/espocrm/command.sh upgrade again?

- Yes, you should split these data sets as shown in the documentation (every section mentioned). After, you can run the upgrade command only with a backup performed!

I see there this an optional step in the instructions to remove obsolete files which appears (amongst other things) to delete command.php, which is called from command,sh - should I not do this?

- Since this step is flagged as optional, you can try to skip this one.

I also notice that image listed under the espocrm section is espocrm/espocrm:fpm but the instructions just show espocrm/espocrm is there any need to change that.

- Install script using slightly different logic from direct Docker installation script while creating a working build (afaik), so image names like "espocrm/espocrm:fpm" are intended for your case and should not be changed.

All questions above are reasonable, so don't worry about asking further ones. Hope everything went/ will go smoothly on your side.

[Gomez]

heint

Well I had a few false starts with a couple of typo's in docker-compose.yml but then I thought I had cracked it as /var/www/espocrm/command.sh upgrade completed, apparently with no errors.

Alas I spoke to soon as now when I try and log in all I get is a 404 Not Found error from nginx

Status returns

CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
9218c64d9292 nginx:latest "/docker-entrypoint.…" 21 minutes ago Up 18 minutes 0.0.0.0:8081->80/tcp, [::]:8081->80/tcp espocrm-nginx
198976a9bbf8 espocrm/espocrm:fpm "docker-websocket.sh" 21 minutes ago Restarting (0) 27 seconds ago espocrm-websocket
f02078e6cdd0 espocrm/espocrm:fpm "docker-daemon.sh" 21 minutes ago Restarting (0) 29 seconds ago espocrm-daemon
6dd404d695f9 espocrm/espocrm:fpm "docker-entrypoint.s…" 21 minutes ago Restarting (1) 24 seconds ago espocrm
d5a20315ee10 mariadb:latest "docker-entrypoint.s…" 21 minutes ago Up 18 minutes (healthy) 3306/tcp espocrm-db

Logs just shows

info: Running "install" action.

about 20 times

Help! I don't know where to start trying to fix this.