A little advanced job scheduling with AP Scheduler

Reason, purpose and motive

I have reached the limit of running a job about once every 10 minutes with cron.
__ If the previous process takes more than 10 minutes, multiple startups will occur. __ I want to prevent this.
I thought I could use some locking mechanism, but I don't want to put it in the job code.
It's annoying that the job depends on the target lock mechanism.
__ Tests are hard and cause bugs. __
Since it is a python job that is running, is there anything written in python if possible?

Searching for a job scheduler that can prevent multiple startups

There was AP Shceduler!
Implemented in python.
Prevents multiple startup.
You can also specify permission for multiple startup and how many are OK.
Cron-like job scheduling is possible.
It can be started once at a specified time like the at command.
Job scheduling in seconds, which is finer than cron, is also possible.
Click here for the original document
http://pythonhosted.org/APScheduler/index.html

Installation

With pip or easy_install.

$ easy_install apscheduler
Or
$ pip install apscheduler

Let's write a job scheduler

Try to implement a scheduler of the type that starts by specifying the number of seconds in the interval.

import apscheduler.scheduler as ApSched

def job_function():
    print "Hello World"

sched = ApSched.Scheduler(standalone=True,coalesce=True)
sched.add_interval_job(job_function, seconds=10)
sched.start()

The standalone option is False by default. If set to True, the scheduler itself will run demonic.

The result of trying it with ipython.

In [18]: 2013-11-26 16:12:32.680493
Hello World
In [18]: 2013-11-26 16:12:42.680962
Hello World
In [18]: 2013-11-26 16:12:52.680975
Hello World

In addition, the interval time is calculated by properly omitting the program execution time.

Experiment to prevent multiple activation

from apscheduler.scheduler import Scheduler
import datetime
import time

def job_function_13sec():
    print datetime.datetime.today()
    time.sleep(13)
    print "Hello World"

sched = Scheduler(standalone=True,coalesce=True)
sched.add_interval_job(job_function_13sec, seconds=10)
sched.start()

result.

In [46]: sched.start()
2013-11-26 16:36:10.818178
WARNING:apscheduler.scheduler:Execution of job "job_function_13sec (trigger: interval[0:00:10], next run at: 2013-11-26 16:36:30.817553)" skipped: maximum number of running instances reached (1)
Hello World
2013-11-26 16:36:30.818776
WARNING:apscheduler.scheduler:Execution of job "job_function_13sec (trigger: interval[0:00:10], next run at: 2013-11-26 16:36:40.817553)" skipped: maximum number of running instances reached (1)
Hello World
2013-11-26 16:36:50.818992
WARNING:apscheduler.scheduler:Execution of job "job_function_13sec (trigger: interval[0:00:10], next run at: 2013-11-26 16:37:10.817553)" skipped: maximum number of running instances reached (1)
Hello World
2013-11-26 16:37:10.818815
WARNING:apscheduler.scheduler:Execution of job "job_function_13sec (trigger: interval[0:00:10], next run at: 2013-11-26 16:37:30.817553)" skipped: maximum number of running instances reached (1)
Hello World
2013-11-26 16:37:30.818828

As a job option __ By default max_run is 1__, which prevents multiple launches.

afterwards

Daemonized with supervisord to run all the time like cron.
http://qiita.com/yushin/items/15f4f90c5663710dbd56
Rest assured that the process will automatically restart even if it dies.
But I want to skip the email if it crashes ... Wow. It seems that you can go with functions such as event listeners.