System Metrics/Stats Library for Linux
- 2010-2013 Corey Goldberg
- Dev Home: https://github.com/cgoldberg/linux-metrics
- PyPI: https://pypi.python.org/pypi/linux-metrics
- Free Open Source : MIT License
linux-metrics is a Python package containing modules for getting OS metrics on systems running the Linux kernel. It is a pure python library with no external dependencies.
Basic stats for major subsystems are provided (Processor/CPU, Disk, Memory, Network).
- pip install linux-metrics
- Python 2.6/2.7/3.1/3.2/3.3/PyPy
- Linux 2.6+
print number of processes running:
from linux_metrics import cpu_stat print cpu_stat.procs_running()
print CPU utilization every 5 secs:
>>> from linux_metrics import cpu_stat >>> >>> while True: ... cpu_pcts = cpu_stat.cpu_percents(5) ... print 'cpu utilization: %.2f%%' % (100 - cpu_pcts['idle']) ... cpu utilization: 0.70% cpu utilization: 0.50% cpu utilization: 24.80% cpu utilization: 20.89% cpu utilization: 40.04%
* linux_metrics * cpu_stat * cpu_times() * cpu_percents(sample_duration=1) * procs_running() * procs_blocked() * load_avg() * cpu_info() * disk_stat * disk_busy(device, sample_duration=1) * disk_reads_writes(device) * disk_usage(path) * disk_reads_writes_persec(device, sample_duration=1) * mem_stat * mem_stats() * net_stat * rx_tx_bytes(interface) * rx_tx_bits(interface) * rx_tx_dump(interface)
linux-metrics package contains an example script:
You can run the included unit tests and verify all cases pass in your environment:
$ nosetests
Note: you may need to adjust the configuration of the unit tests to match your environment. They are set by default to use:
DISK_DEVICE = 'sda1' NETWORK_INTERFACE = 'eth0'