java-gc 0.2.0

Trigger a full Java garbage collection

A command to trigger a full java garbage collection via the jcmd tool. On systems which use the systemd init system, you can specify the service name as the program argument. The -r option can also be used on systemd based systems to allow have this utility attempt to automatically restart the service if the jcmd garbage collection fails.

Requirements

• Tested on python 2.7 and python 3

• No 3rd party libraries required

Installation

Using pip is the preferred method to install this tool, and will install the package from pypi. pip install java-gc

It can also be installed via pip from a local copy of the source repo, pip install . from the repo directory should do the trick. Running make install should also work as well.

Usage

usage: java-gc [-h] [-u USER] [-t T] [-r]

[-l {debug,info,warning,error,critical}] [-v]

proc

Trigger a full GC on a java process

positional arguments:

proc The service name, or pid, to trigger the GC on

optional arguments:

-h, –help show this help message and exit

-u USER, –user USER User owning the process (default to current user)

-t T, –timeout T seconds to wait for GC before raising error

-r, –restart restart the service if timeout occurred, else exit

with error

-l {debug,info,warning,error,critical}, –loglevel {debug,info,warning,error,critical}

set log level

-v, –version print program version and exit

Contributing

The usual github model for forking the repo and creating a pull request is the preferred way to contribute to this tool. Bug fixes, enhancements, doc updates, translations are always welcomed.

Building from source

A Makefile has been included in the repository to handle the steps of creating the package and uploading it to pypi. If you don’t have the make utility installed, the contents of the Makefile should be instructive (and simple) enough to execute manually.

The following targets are available in the Makefile:

• package - the default target, calls the setup.py script to create the package to upload to pypi.

• upload - depends on the package target, and uploads the generated package archive to pypi.

• install - use pip to install the package locally

• clean - clean up the artifacts left by the package step.

• distclean - depends on the clean target, and additionally cleans up misc. files.