Follow Glances on Twitter: @nicolargo or @glances_system
Glances is a cross-platform curses-based system monitoring tool written in Python.
python 2.7,>=3.3
(tested with version 2.7, 3.3, 3.4, 3.5)psutil>=2.0.0
setuptools
Optional dependencies:
bottle
(for Web server mode)py3sensors
(for hardware monitoring support) [Linux-only]hddtemp
(for HDD temperature monitoring support) [Linux-only]batinfo
(for battery monitoring support) [Linux-only]pymdstat
(for RAID support) [Linux-only]pysnmp
(for SNMP support)zeroconf
(for the autodiscover mode)netifaces
(for the IP plugin)influxdb
(for the InfluxDB export module)statsd
(for the StatsD export module)pystache
(for the action script feature)docker-py
(for the Docker monitoring support) [Linux-only]matplotlib
(for graphical/chart support)pika
(for the RabbitMQ/ActiveMQ export module)bernhard
(for the Riemann export module)py-cpuinfo
(for the Quicklook CPU info module)scandir
(for the Folders plugin) [Only for Python < 3.5]
Note for Python 2.6 users
Since version 2.7, Glances no longer support Python 2.6. Please upgrade to at least Python 2.7/3.3+ or downgrade to Glances 2.6.2 (latest version with Python 2.6 support).
Note for CentOS Linux 6 and 7 users
Python 2.7, 3.3 and 3.4 are now available via SCLs. See: https://lists.centos.org/pipermail/centos-announce/2015-December/021555.html.
Several method to test/install Glances on your system. Choose your weapon !
To install both dependencies and latest Glances production ready version (aka master branch), just enter the following command line:
curl -L https://bit.ly/glances | /bin/bash
or
wget -O- https://bit.ly/glances | /bin/bash
Note: Only supported on some GNU/Linux distributions. If you want to support other distributions, please contribute to glancesautoinstall.
Glances is on PyPI
. By using PyPI, you are sure to have the latest
stable version.
To install, simply use pip
:
pip install glances
Note: Python headers are required to install psutil. For example, on Debian/Ubuntu you need to install first the python-dev package. For Fedora/CentOS/RHEL install first python-devel package. For Windows, just install PsUtil from the binary installation file.
You can also install the following libraries in order to use optional features (like the Web interface, exports modules, sensors...):
pip install bottle requests batinfo https://bitbucket.org/gleb_zhulik/py3sensors/get/tip.tar.gz zeroconf netifaces pymdstat influxdb elasticsearch potsdb statsd pystache docker-py pysnmp pika py-cpuinfo bernhard cassandra scandir
To upgrade Glances to the latest version:
pip install --upgrade glances
If you need to install Glances in a specific user location, use:
export PYTHONUSERBASE=~/mylocalpath
pip install --user glances
A Glances container is available. It will include the latest development HEAD version. You can use it to monitor your server and all your others containers !
Get the Glances container:
docker pull nicolargo/glances
Run the container in console mode:
docker run -v /var/run/docker.sock:/var/run/docker.sock:ro --pid host -it docker.io/nicolargo/glances
Run the container in Web server mode (notice the GLANCES_OPT environment variable setting parameters for the glances startup command) :
docker run -d -p 61208:61208 -e GLANCES_OPT="-w" -v /var/run/docker.sock:/var/run/docker.sock:ro --pid host docker.io/nicolargo/glances
Additionally, If you want to use your own glances.conf file, you can create your own Dockerfile:
FROM nicolargo/glances
COPY glances.conf /glances/conf/glances.conf
CMD python -m glances -C /glances/conf/glances.conf $GLANCES_OPT
Alternatively, you can specify something along the same lines with docker run options:
docker run -v ./glances.conf:/glances/conf/glances.conf -v /var/run/docker.sock:/var/run/docker.sock:ro --pid host -it docker.io/nicolargo/glances
Where ./glances.conf is a local directory containing your glances.conf file.
At the moment, packages exist for the following GNU/Linux distributions:
- Arch Linux
- Debian
- Fedora/CentOS/RHEL
- Gentoo
- Slackware (SlackBuild)
- Ubuntu
- Void Linux
So you should be able to install it using your favorite package manager. Be aware that the Glances version can not be the latest one using this method.
To install the binary package:
# pkg install py27-glances
To install Glances from ports:
# cd /usr/ports/sysutils/py-glances/
# make install clean
OS X users can install Glances using Homebrew
or MacPorts
.
$ brew install python
$ pip install glances
$ sudo port install glances
Install Python for Windows (Python 2.7.9+ and 3.4+ ship with pip) and then just:
$ pip install glances
You can also install the following libraries in order to use optional features (like the Web interface, exports modules, sensors...):
pip install bottle requests zeroconf netifaces influxdb elasticsearch potsdb statsd pystache pysnmp pika py-cpuinfo bernhard cassandra scandir
To install Glances from source:
$ wget https://github.com/nicolargo/glances/archive/vX.Y.tar.gz -O - | tar xz
$ cd glances-*
# python setup.py install
Note: Python headers are required to install psutil. For example, on Debian/Ubuntu you need to install first the python-dev package.
An awesome Chef
cookbook is available to monitor your infrastructure: https://supermarket.chef.io/cookbooks/glances (thanks to Antoine Rouyer)
You can install Glances using Puppet
: https://github.com/rverchere/puppet-glances
For the standalone mode, just run:
$ glances
For the Web server mode, run:
$ glances -w
and enter the URL http://<ip>:61208
in your favorite web browser.
For the client/server mode, run:
$ glances -s
on the server side and run:
$ glances -c <ip>
on the client one.
You can also detect and display all Glances servers available on your network or defined in the configuration file:
$ glances --browser
and RTFM, always.
For complete documentation have a look at the readthedocs website.
If you have any question (after RTFM!), please post it on the official Q&A forum.
Glances can export stats to: CSV
file, InfluxDB
, Cassandra
,
OpenTSDB
, StatsD
, ElasticSearch
, RabbitMQ
and Riemann
server.
If you want to contribute to the Glances project, read this wiki page.
There is also a chat dedicated to the Glances developers:
Nicolas Hennion (@nicolargo) <[email protected]>
LGPL. See COPYING
for more details.