In my Laravel application, I have created a folder .ebextensions and it has the configuration to install supervisor in the EC2 instance.
When I deploy the application for the first time and the instance gets created, everything works fine. Supervisor gets installed.
But when the instance scales and a new EC2 gets spawned, it doesn't take the same configuration. I need to install supervisor manually on the newer instance.
Is there a way, where the newer instances would take the configuration from .ebextensions and run it in the similar way it did the first time?
This is the structure of the .ebextensions folder
.ebextensions
- supervisor
- setup.sh
- supervisor_laravel.conf
- supervisord.conf
- supervisor.config
setup.sh
#!/bin/bash
echo "Supervisor - starting setup"
. /opt/elasticbeanstalk/deployment/env
if [ ! -f /usr/bin/supervisord ]; then
echo "installing supervisor"
easy_install supervisor
else
echo "supervisor already installed"
fi
if [ ! -d /etc/supervisor ]; then
mkdir /etc/supervisor
echo "create supervisor directory"
fi
if [ ! -d /etc/supervisor/conf.d ]; then
mkdir /etc/supervisor/conf.d
echo "create supervisor configs directory"
fi
. /opt/elasticbeanstalk/deployment/env && cat .ebextensions/supervisor/supervisord.conf > /etc/supervisor/supervisord.conf
. /opt/elasticbeanstalk/deployment/env && cat .ebextensions/supervisor/supervisord.conf > /etc/supervisord.conf
. /opt/elasticbeanstalk/deployment/env && cat .ebextensions/supervisor/supervisor_laravel.conf > /etc/supervisor/conf.d/supervisor_laravel.conf
if ps aux | grep "[/]usr/bin/supervisord"; then
echo "supervisor is running"
else
echo "starting supervisor"
/usr/bin/supervisord
fi
/usr/bin/supervisorctl reread
/usr/bin/supervisorctl update
echo "Supervisor Running!"
yum -y install http://cbs.centos.org/kojifiles/packages/beanstalkd/1.9/3.el7/x86_64/beanstalkd-1.9-3.el7.x86_64.rpm
if ps aux | grep "[/]usr/bin/beanstalkd"; then
echo "beanstalkd is running"
else
echo "starting beanstalkd"
/bin/systemctl start beanstalkd.service
fi
echo "Beanstalkd Running..."
supervisor_laravel.conf
[program:worker]
process_name=%(program_name)s_%(process_num)02d
command=/usr/bin/php /var/www/html/artisan queue:work --tries=3 --timeout=0
autostart=true
autorestart=true
stopasgroup=true
killasgroup=true
;user=forge
numprocs=3
redirect_stderr=true
stderr_logfile=/var/log/supervisor_laravel.err.log
stdout_logfile=/var/log/supervisor_laravel.out.log
supervisor.conf
[unix_http_server]
file=/tmp/supervisor.sock ; (the path to the socket file)
[supervisord]
logfile=/tmp/supervisord.log ; (main log file;default $CWD/supervisord.log)
logfile_maxbytes=50MB ; (max main logfile bytes b4 rotation;default 50MB)
logfile_backups=10 ; (num of main logfile rotation backups;default 10)
loglevel=info ; (log level;default info; others: debug,warn,trace)
pidfile=/tmp/supervisord.pid ; (supervisord pidfile;default supervisord.pid)
nodaemon=false ; (start in foreground if true;default false)
minfds=1024 ; (min. avail startup file descriptors;default 1024)
minprocs=200 ; (min. avail process descriptors;default 200)
environment=SYMFONY_ENV=prod
[rpcinterface:supervisor]
supervisor.rpcinterface_factory = supervisor.rpcinterface:make_main_rpcinterface
[supervisorctl]
serverurl=unix:///tmp/supervisor.sock ; use a unix:// URL for a unix socket
[include]
files = /etc/supervisor/conf.d/*.conf
[inet_http_server]
port = 9000
username = user
password = pw
supervisor.config
container_commands:
01_install_supervisor:
command: ".ebextensions/supervisor/setup.sh"
I have found the answer. But slightly in a different way.
Initially, I had all the files in .ebextensions folder, I moved some of the files into a different folder called "awsconfig" (The name doesn't matter), and moved some of the commands of setup.sh file to .platform/hooks/postdeploy and gave the permission as chmod +x to setup.sh
Important
AWS cleans up the .ebextensions folder after they are executed. I read that here
I did this because supervisor was getting installed in the newer instance (the instance which gets created after scaling), but the configuration files weren't getting copied because there was no .ebextensions folder in the newer instance
To copy the configuration files after the instance gets deployed, I used platform hooks postdeploy.
You can read about platform-hooks-postdeploy here
This is the structure I have now
These are the changes in setup.php file which is split into two files in different folders
.ebextensions/supervisor/setup.sh
.platform/hooks/postdeploy/setup.sh
The content of all other files remains the same as what I have posted in the question.