A two part tool for monitoring callstack statistics of CherryPy appliations and analysing/displaying these statistics.
The client is a small-footprint package to be imported into your application, the configuration you supply will dictate which CherryPy handlers, which functions and which application database will be wrapped for callstack and SQL profiling. The configuration also controls how often these collected stats are flushed and where they are pushed to.
It works by subscribing to the CherryPy engine at startup, wrapping listed functions, wrapping database connection objects and activating custom CherryPy tools which wrap cherrypy handlers. It then collects cProfile callstack data and SQL call durations from them. These stats are written to temporary buffers and flushed periodically.
Run the setup of the client by running the following in the shell/cmd terminal:
setup.py install
Then insert the following lines into your application:
import cherry_pyformance
cherry_pyformance.initialise()
Then copy and configure the default config JSON file into the same directory and rename to 'cherrypyformance_config.cfg'.
- Python 2.6/7
- CherryPy
The server is a CherryPy application which accepts stats pushed by the cherry_pyformance client storing them for analysis and browsing.
Go to the server directory. Copy 'server_config.cfg.template' to 'server_config.cfg'. Enter your database and server details into the config file. Copy 'alembic.ini.template' to 'alembic.ini'. Change the 'sqlalchemy.url' line, modifying the argument to have your database's username/password where it says 'username' and 'password'. Then run
python stats_server.py
The server UI then runs on the host and port specified in the config. Your server should upgrade the database automatically using Alembic.
- Python 2.6/7
- CherryPy
- Mako
- SQLAlchemy
- Alembic
- SQLParse