cron with secondary container not working

[friendica]

cron with secondary container not working

[zauberstuhl]

FYI it worked for me after I added:

[system]
pidfile = "/var/run/friendica.pid"

to config/local.ini.php

[zauberstuhl]

@noplanman the cron.sh starts a daemon script. if you want everything in one container it might make sense to alter the entrypoint and start the workers permanently via:

php /var/www/html/bin/daemon.php -f start

Then you have no host dependencies.

[nupplaphil]

• The Zombie issue is resolved by using gosu and tini, changed #162 - it never happened again since this change
• The pidfile is now part of the core config, so no need to add it extra (https://github.com/friendica/docker/blob/stable/.config/zz-docker.config.php#L13)

So I'm closing this issue :)

[nupplaphil]

Did you use some of the .examples as baseline? Can you paste your docker-compose.yml if possible or the startup environment variables.

Is the "normal" Friendica node working? Can you see the page? Is Friendica already installed?

this one is my configuration, which is working.

.env - File

# ------------------------------                                   
# friendica configuration                                          
# ------------------------------                                   
# example.org is _not_ a valid hostname, use a fqdn here.          
HOSTNAME=friendica.philipp.info                                    
# ------------------------------                                   
# SQL database configuration                                       
# ------------------------------                                   
DBUSER=********
DBDATA=********                                         
DBHOST=db
DBPORT=3306
# Please use long, random alphanumeric strings (A-Za-z0-9)
DBPASS=***********
                                                                   
SITENAME="Philipp.info Friendica"

docker-compose.yml

version: '2.1'      
services:           
   cron:                                                       
     image: friendica/server:dev-fpm-alpine                    
     restart: always                                           
     volumes:                                                  
       - friendica-vol-1:/var/www/html                         
     environment:                                              
       - SITENAME=${SITENAME}                                  
       - SMTP=smtp.philipp.info                                
     depends_on:                                               
       - db                                                    
       - redis                                                 
     entrypoint: /cron.sh                                      
     hostname: friendica.philipp.info                          
     logging:                                                  
       driver: json-file                                       
       options:                                                
         max-size: "10m"                                       
         max-file: "3"                                         

[noplanman]

I've just opted to set a cronjob in the host machine, instead of a second container:
*/10 * * * * root /usr/bin/docker exec -u www-data friendica php bin/worker.php

Not sure how "good" or "bad" this is, but it's been working super smooth since the very beginning

[iannuzzelli]

I had a very similar issue and ended up creating the cronjob in the host as suggested by @noplanman , although not nice as it creates a dependency in case of future scaling needs.

In my current setup (using friendica:2020.03-apache image), the cron container, despite being passed the same env file of the friendica app, doesn't start the daemon, it just hangs reporting this warning

[WARN  tini (253)] Tini is not running as PID 1 and isn't registered as a child subreaper.
Zombie processes will not be re-parented to Tini, so zombie reaping won't work.
To fix the problem, use the -s option or set the environment variable TINI_SUBREAPER to register Tini as a child subreaper, or run Tini as PID 1.

then it tries to send the error notification via email, using the wrong SMTP (I believe that's because the SMTP env variables are not parsed by cron.sh, which is probably a separate issue)