def process(name, **kwargs):
"""
Require a supervisor process
"""
from fabtools import require
require.deb.package('supervisor')
require.service.started('supervisor')
# Set default parameters
params = {}
params.update(kwargs)
params.setdefault('autorestart', 'true')
params.setdefault('redirect_stderr', 'true')
# Build config file from parameters
lines = []
lines.append('[program:%(name)s]' % locals())
for key, value in sorted(params.items()):
lines.append("%s=%s" % (key, value))
# Upload config file
filename = '/etc/supervisor/conf.d/%(name)s.conf' % locals()
with watch(filename, callback=update_config, use_sudo=True):
require.file(filename, contents='\n'.join(lines), use_sudo=True)
# Start the process if needed
if process_status(name) == 'STOPPED':
start_process(name)
def process(name, **kwargs):
"""
Require a supervisor process to be running.
Keyword arguments will be used to build the program configuration
file. Some useful arguments are:
- ``command``: complete command including arguments (**required**)
- ``directory``: absolute path to the working directory
- ``user``: run the process as this user
- ``stdout_logfile``: absolute path to the log file
You should refer to the `supervisord documentation`_ for the
complete list of allowed arguments.
.. note:: the default values for the following arguments differs from
the supervisord defaults:
- ``autorestart``: defaults to ``true``
- ``redirect_stderr``: defaults to ``true``
Example::
from fabtools import require
require.supervisor.process('myapp',
command='/path/to/venv/bin/myapp --config production.ini --someflag',
directory='/path/to/working/dir',
user='******',
stdout_logfile='/path/to/logs/myapp.log',
)
.. _supervisord documentation: http://supervisord.org/configuration.html#program-x-section-values
"""
from fabtools import require
require.deb.package("supervisor")
require.service.started("supervisor")
# Set default parameters
params = {}
params.update(kwargs)
params.setdefault("autorestart", "true")
params.setdefault("redirect_stderr", "true")
# Build config file from parameters
lines = []
lines.append("[program:%(name)s]" % locals())
for key, value in sorted(params.items()):
lines.append("%s=%s" % (key, value))
# Upload config file
filename = "/etc/supervisor/conf.d/%(name)s.conf" % locals()
with watch(filename, callback=update_config, use_sudo=True):
require.file(filename, contents="\n".join(lines), use_sudo=True)
# Start the process if needed
if process_status(name) == "STOPPED":
start_process(name)
def process(name, **kwargs):
"""
Require a supervisor process to be running.
Keyword arguments will be used to build the program configuration
file. Some useful arguments are:
- ``command``: complete command including arguments (**required**)
- ``directory``: absolute path to the working directory
- ``user``: run the process as this user
- ``stdout_logfile``: absolute path to the log file
You should refer to the `supervisor documentation`_ for the
complete list of allowed arguments.
.. note:: the default values for the following arguments differs from
the ``supervisor`` defaults:
- ``autorestart``: defaults to ``true``
- ``redirect_stderr``: defaults to ``true``
Example::
from fabtools import require
require.supervisor.process('myapp',
command='/path/to/venv/bin/myapp --config production.ini --someflag',
directory='/path/to/working/dir',
user='******',
stdout_logfile='/path/to/logs/myapp.log',
)
.. _supervisor documentation: http://supervisord.org/configuration.html#program-x-section-values
"""
from fabtools import require
require.deb.package('supervisor')
require.service.started('supervisor')
# Set default parameters
params = {}
params.update(kwargs)
params.setdefault('autorestart', 'true')
params.setdefault('redirect_stderr', 'true')
# Build config file from parameters
lines = []
lines.append('[program:%(name)s]' % locals())
for key, value in sorted(params.items()):
lines.append("%s=%s" % (key, value))
# Upload config file
filename = '/etc/supervisor/conf.d/%(name)s.conf' % locals()
with watch(filename, callback=update_config, use_sudo=True):
require.file(filename, contents='\n'.join(lines), use_sudo=True)
# Start the process if needed
if process_status(name) == 'STOPPED':
start_process(name)
def test_uwsgi_is_started(now):
for i in range(1, 30):
status = supervisor.process_status('uwsgi_{}'.format(now))
if status == 'RUNNING':
break
time.sleep(1)
testing_file = '/tmp/test_uwsgi.py'
if files.is_file(testing_file):
files.remove(testing_file)
put('files/test_uwsgi.py', '/tmp/')
require.python.package('six', use_sudo=True)
output = run('python {} {} {} aa'.format(testing_file, env.uwsgi_socket_api(now),
'{}/ads/'.format(env.conf_api.SERVER_NAME)))
assert '"message"' in output
from test_api import test_api
test_api(testing_file, env.uwsgi_socket_api(now), env.conf_api.SERVER_NAME)
def process(name, **kwargs):
"""
Require a supervisor process to be running.
Keyword arguments will be used to build the program configuration
file. Some useful arguments are:
- ``command``: complete command including arguments (**required**)
- ``directory``: absolute path to the working directory
- ``user``: run the process as this user
- ``stdout_logfile``: absolute path to the log file
You should refer to the `supervisor documentation`_ for the
complete list of allowed arguments.
.. note:: the default values for the following arguments differs from
the ``supervisor`` defaults:
- ``autorestart``: defaults to ``true``
- ``redirect_stderr``: defaults to ``true``
Example::
from fabtools import require
require.supervisor.process('myapp',
command='/path/to/venv/bin/myapp --config production.ini --someflag',
directory='/path/to/working/dir',
user='******',
stdout_logfile='/path/to/logs/myapp.log',
)
.. _supervisor documentation: http://supervisord.org/configuration.html#program-x-section-values
"""
from fabtools.require import file as require_file
from fabtools.require.deb import package as require_deb_package
from fabtools.require.rpm import package as require_rpm_package
from fabtools.require.arch import package as require_arch_package
from fabtools.require.service import started as require_started
family = distrib_family()
if family == 'debian':
require_deb_package('supervisor')
require_started('supervisor')
elif family == 'redhat':
require_rpm_package('supervisor')
require_started('supervisord')
elif family == 'arch':
require_arch_package('supervisor')
require_started('supervisord')
else:
raise UnsupportedFamily(supported=['debian', 'redhat', 'arch'])
# Set default parameters
params = {}
params.update(kwargs)
params.setdefault('autorestart', 'true')
params.setdefault('redirect_stderr', 'true')
# Build config file from parameters
lines = []
lines.append('[program:%(name)s]' % locals())
for key, value in sorted(params.items()):
lines.append("%s=%s" % (key, value))
# Upload config file
if family == 'debian':
filename = '/etc/supervisor/conf.d/%(name)s.conf' % locals()
elif family == 'redhat':
filename = '/etc/supervisord.d/%(name)s.ini' % locals()
elif family == 'arch':
filename = '/etc/supervisor.d/%(name)s.ini' % locals()
with watch(filename, callback=update_config, use_sudo=True):
require_file(filename, contents='\n'.join(lines), use_sudo=True)
# Start the process if needed
if process_status(name) == 'STOPPED':
start_process(name)
def process(name, use_pip=False, **kwargs):
"""
Require a supervisor process to be running. Installs supervisor
from the default system package manager unless ``use_pip`` is
truthy.
Keyword arguments will be used to build the program configuration
file. Some useful arguments are:
- ``command``: complete command including arguments (**required**)
- ``directory``: absolute path to the working directory
- ``user``: run the process as this user
- ``stdout_logfile``: absolute path to the log file
You should refer to the `supervisor documentation`_ for the
complete list of allowed arguments.
.. note:: the default values for the following arguments differs from
the ``supervisor`` defaults:
- ``autorestart``: defaults to ``true``
- ``redirect_stderr``: defaults to ``true``
Example::
from fabtools import require
require.supervisor.process('myapp',
command='/path/to/venv/bin/myapp --config production.ini --someflag',
directory='/path/to/working/dir',
user='******',
stdout_logfile='/path/to/logs/myapp.log',
)
.. _supervisor documentation: http://supervisord.org/configuration.html#program-x-section-values
"""
from fabtools.require import file as require_file
from fabtools.require.python import package as require_python_package
from fabtools.require.deb import package as require_deb_package
from fabtools.require.rpm import package as require_rpm_package
from fabtools.require.arch import package as require_arch_package
from fabtools.require.service import started as require_started
# configure installation. override default package installation w/ use_pip
family = distrib_family()
if family == 'debian':
require_package = require_deb_package
package_name = 'supervisor'
daemon_name = 'supervisor'
filename = '/etc/supervisor/conf.d/%(name)s.conf' % locals()
elif family == 'redhat':
require_package = require_rpm_package
package_name = 'supervisor'
daemon_name = 'supervisord'
filename = '/etc/supervisord.d/%(name)s.ini' % locals()
elif family == 'arch':
require_package = require_arch_package
package_name = 'supervisor'
daemon_name = 'supervisord'
filename = '/etc/supervisor.d/%(name)s.ini' % locals()
else:
raise UnsupportedFamily(supported=['debian', 'redhat', 'arch'])
if use_pip:
require_package = require_python_package
package_name = 'supervisor'
# install supervisor and make sure its started
require_package(package_name)
require_started(daemon_name)
# Set default parameters
params = {}
params.update(kwargs)
params.setdefault('autorestart', 'true')
params.setdefault('redirect_stderr', 'true')
# Build config file from parameters
lines = []
lines.append('[program:%(name)s]' % locals())
for key, value in sorted(params.items()):
lines.append("%s=%s" % (key, value))
# Upload config file
with watch(filename, callback=update_config, use_sudo=True):
require_file(filename, contents='\n'.join(lines), use_sudo=True)
# Start the process if needed
if process_status(name) == 'STOPPED':
start_process(name)
def restart(config):
app_name = _get_app_name(config['GIT_URL'])
supervisor.restart_process(app_name)
time.sleep(1)
print(app_name, 'status:', supervisor.process_status(app_name))
def supervisor_status(name):
return supervisor.process_status(name)
