---

DS Scheduler

_______________________________________________________________________________ A centralized 'cron' type scheduling system for UNIX/Linux. It has a web interface for managing, monitoring and scheduling jobs and commands in a multi-host environment. Can execute jobs based on event triggers. Secure communications between master servers and clients. Clusterable: Master can automatically fail over so any number of secondary masters. Ajax web interface for better monitoring and execution feedback. Graphing/Reporting: Visualize the running duration and timing of your jobs.

Features:

• Can be used as a drop-in replacement for cron.
• Can execute jobs based on event triggers.
• Secure communications between master servers and clients.
• Cluster-able: Master can automatically fail over to any number of secondary masters.
• Ajax web interface for better monitoring and execution feedback.
• Graphing/Reporting: Visualize the running duration and timing of your jobs.
• Run Jobs based on File Triggers
• Database will upgrade itself.
• Web interface will indicate if the master is running or not.

Installation:

Requirements:

• APSchedler - available at http://packages.python.org/APScheduler

• Apache Web server

• Python >= 2.6

Point your web server to the 'web' directory that is under scheduler/. If you installed in /opt/scheduler, then an example apache entry would be:

''' Alias /scheduler/ /opt/scheduler/web/ <Directory /opt/scheduler/web/> AuthType Basic AuthName "DS_SChed" AuthUserFile /opt/scheduler/.htpasswd require valid-user '''

Secure your site by using SSL and htpasswd. You can run root processes with the scheduler to restricting access is highly recommended.

Creating an htpasswd file:

htpasswd -c /opt/scheduler/.htpasswd

Create the Database

Create a mysql user and a blank DB in MySQL. Make sure to update the sched_config.py and the dbconfig.php files to reflect the database user, host, and passwd. Note for version 0.5 and later:

The sched_master application will create and upgrade the database tables. The goal here it to avoid doing any DB maintenance to allow for very easy upgrades.

Screenshots: Web interface:

Graphs:

Admin Console (New in 0.7):

Those that prefer the Command Line will be happy that I am working on a terminal based console for DS-scheduler. Just run sched_console:

Hit the H key to bring up the key commands. Sched_admin (New in 0.7):

Managing jobs are easy to script now with this command line interface:

Usage: sched_admin [OPTIONS]

Options: -h , --help Print this info -l, --list List jobs in the system -r, --run [ id ] Run job id immediately -k, --kill [ id ] Kill running job -a, --add [ id ] Add a new job --cron --name --host --user --depends [ id ] --mode [ ON_SUCCESS | ON_FAILURE ] -d, --delete [ id ] Delete job completely -e, --enable [ id ] Enable a disabled job -x, --disable [ id ] Disable job from running -q, --quiet Quiet mode ( good for scripting )

TO DO:

Not every feature on the web interface works yet. Anyone who knows web development could probably do better. Patches are welcome. Update: Most web features now work. Testing needed.
Need to add ON_NEWFILE trigger.  Done
Need to add ON_FILECHANGE trigger.
Need better reporting/graphing flexibility.
Need Translations

Database:

For those who want to contribute to the development, Here is an outline of the database:

Table: Jobs

name varchar(255) Human readable job name host varchar(30) Host that job will run on user varchar(10) user that job will run as start_time datetime The last time the job started end_time datetime The last tim the job ended depends int(11) Does this job depend on another? depend_mode int(11) See the dep_mode table reoccur varchar(30) 'cron' type notation status int(11) The running status rc int(11) The exit code of the job command varchar(300) The actual command to run update_flag tinyint(1) Trigger the scheduler to re-read the job comment text Human readable comments Log_Retention int How long in days to keep run logs

Table: dep_mode

For managing dependant modes

mode_id int(11) Unique ID mode varchar(25) Description of mode (i.e. ON_FAILURE)

Table: Results

For storing output and times for reporting

id int(11) Unique ID start_date datetime Time the job started end_date datetime Time the job exited output text Command/Job output text uid varchar(32) Unique identifier for each job instance

Table: Events

Table: db_version

This table is for keeping track of the database structure.

id int(11) db_version int(11)

Contributing

• Fork it
• Create your feature branch (git checkout -b my-new-feature)
• Commit your changes (git commit -am 'Add some feature')
• Push to the branch (git push origin my-new-feature)
• Create new Pull Request

id

int(11)

AUTO_INCREMENT unique job ID