introduction

SeAT is able to query a supervisor instance for worker statuses and other information. This information can then be displayed in the Queue Status screen.

requirements

Supervisor monitoring needs Supervisor version 3.0 or later. Check your supervisor version with supervisord -v. If you are running CentOS 6, please refer to the updating supervisor on centos 6 section below before continuing.

the config

A bit of setup work is needed in order to have your SeAT setup query a supervisor instance. The basic idea to get this working is:

• Add a [inet_http_server] configuration block to your supervisor.conf file and restart supervisor.
• Add the username, password, ip and port of the supervisor instance to your SeAT .env configuration file.

1. add supervisor configuration block.

On your server where supervisor is running, open the configuration file /etc/supervisor/supervisord.conf in your favorite text editor. Add the following configuration block to your configuration file:

[inet_http_server]
port=127.0.0.1:9001
username=seat
password=your_secure_password

Obviously, substitute the values with those that correspond with your environment. For most setups, the values in the example should be just fine.

Finally, restart supervisor itself.

2. configure SeAT with the new values.

SeAT itself also needs some configurations added in order for it to know where the supervisor instance is, as well as what password to use when querying for status information.

Again, on your server, find your SeAT installations .env file (usually located at /var/www/seat/.env) and add the following configuration block:

SUPERVISOR_RPC_ADDRESS=127.0.0.1
SUPERVISOR_RPC_USERNAME=seat
SUPERVISOR_RPC_PASSWORD=hd@rjf&@921aJlw0
SUPERVISOR_RPC_PORT=9001
SUPERVISOR_GROUP=seat

The defaults here should also be sufficient but feel free to adjust as needed.

updating supervisor on centos 6

CentOS 6 ships with supervisor 2x by default. In order to get the supervisor monitoring to work, the following steps may be followed to upgrade your supervisor version.

• Install the GhettoForge repository. This can be done with:

# Download and install the latest release
GF=gf-release-6-10.gf.el6.noarch.rpm && curl -O http://mirror.symnds.com/distributions/gf/el/6/gf/x86_64/$GF && yum localinstall -y $GF && rm -f $GF

# Import the GhettoForge signing keys
rpm --import http://mirror.symnds.com/distributions/gf/RPM-GPG-KEY-gf.el6

# Enable the repository
yum install yum-utils -y
yum-config-manager --enable gf-plus

• Next, install Supervisor by running yum install supervisor -y

This will update your local installation to version 3x. Your old configuration will be saved in /etc/supervisord.conf.rpmsave. Supervisor 3 allows you to specify a numprocs setting in a command block instead of having to define multiple instances like you had to do with v2.x. You will need to create a file called seat.ini in /etc/supervisord.d/. This file should look something like this:

[program:seat]
command=/usr/bin/php /var/www/seat/artisan queue:work --queue=high,medium,low,default --tries 1 --timeout=604800
process_name = %(program_name)s-80%(process_num)02d
stdout_logfile = /var/log/seat-80%(process_num)02d.log
stdout_logfile_maxbytes=100MB
stdout_logfile_backups=10
numprocs=6
directory=/var/www/seat
stopwaitsecs=600
user=apache

• Restart supervisor with /etc/init.d/supervisord restart and check that your workers are running again with supervisorctl status.

troubleshooting

In case something goes wrong, you have a few debugging options.

• Obviously, first ensure that supervisor itself is running.
• If the http interface is started on a remote server, ensure that no firewall is blocking TCP traffic to the destination server.
• If you just added the inet_http_server configuration block, restart supervisor and then check back on the web interface.
• Attempt to debug the connection using the tinker example below.
• Hop onto Slack (details in the contact page) for help.

troubleshooting - tinker console

Another option you have to try and figure out why your supervisor integration might not be working would be to attempt to mimic the calls SeAT makes to supervisor and looking for errors.

Don't worry, this is easier than it looks!

First, drop into a tinker shell by running php artisan tinker. This will drop you into a fully booted instance of SeAT:

Next, copy and paste the following instantiation code into the console. This will initialize the object needed to connect to the supervisor inet_http_server using the configuration in your .env config file:

$supervisor = new Supervisor\Supervisor(config('web.supervisor.name'), config('web.supervisor.rpc.address'), config('web.supervisor.rpc.username'), config('web.supervisor.rpc.password'), (int) config('web.supervisor.rpc.port'));

Pasting that should result in output similar to the below screenshot:

Finally, attempt to get the current running processes by running the following code:

$supervisor->getProcesses();

As you can see in my example below, the object is unable to connect to the supervisor inet_http_server. This could indicate a number of things and should hopefully help with troubleshooting any errors that may occur.