https://github.com/niphlod/cs_monitor_plugin

Web2py plugin to monitor the scheduler
https://github.com/niphlod/cs_monitor_plugin

Last synced: 13 days ago
JSON representation

Web2py plugin to monitor the scheduler

• Host: GitHub
• URL: https://github.com/niphlod/cs_monitor_plugin
• Owner: niphlod
• Created: 2013-02-23T17:12:51.000Z (over 11 years ago)
• Default Branch: master
• Last Pushed: 2015-03-09T23:07:27.000Z (over 9 years ago)
• Last Synced: 2024-08-01T22:56:40.557Z (3 months ago)
• Language: Python
• Size: 688 KB
• Stars: 11
• Watchers: 4
• Forks: 5
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
ComfortScheduler Monitor

=================

Since Web2py had its own scheduler, too many peoples asked for an embedded monitoring solution.

ComfortScheduler it's just my way of defining it. 

It bakes a lot of features and relies on a RDBMS, which most of web apps have anyway in their stack.   

Right now the plugin scans the scheduler's tables, and it's a first step towards a more "precise" analyzing solution.

Please try it and report back your findings.

INSTRUCTIONS:

 - download it and merge with your app's code

 - no cache is used ATM, but I'm planning to use that to avoid db pressure. Change in controllers/plugin_cs_monitor.py the lines

    ```python

    sc_cache = cache.ram

    GROUPING_MODE = 'database' # or 'python'

    ANALYZE_CACHE_TIME = 60

    TASKS_SUMMARY_CACHE_TIME = 10

    ```

    to match your preferences

    

    ```python

    sc_cache = cache.ram #your cache backend

    GROUPING_MODE = 'database' # or 'python', to save the db from some heavy queries

    ANALYZE_CACHE_TIME = 60 #how many seconds cache the queries done to analyze the task

    TASKS_SUMMARY_CACHE_TIME = 10 #how many seconds cache the queries for the task summary on the index page

    ```

 

 - the index() function is just a placeholder, every URL is generated with user_signature, so change the permission according to your requirements only in the index() function. 

    ```python

    @auth.requires_login()

    def index():

        return dict()

    ```

    e.g to

    

    ```python

    @auth.requires_membership('superadministrators')

    def index():

        return dict()

    ```

        

TODO:

- [x] Groupings can be done in python too, to avoid db pressure

- [x] analyze_task works scanning scheduler_run results. It should detect automatically if no runs are there and switch groupings.

- [ ] add REST interface ?