ansible/cloud/docker/docker.py

1930 lines
71 KiB
Python
Raw Normal View History

#!/usr/bin/python
# (c) 2013, Cove Schneider
# (c) 2014, Joshua Conner <joshua.conner@gmail.com>
# (c) 2014, Pavel Antonov <antonov@adwz.ru>
2013-10-14 00:07:27 +02:00
#
# This file is part of Ansible,
2013-10-14 00:07:27 +02:00
#
# Ansible is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
2013-10-14 00:07:27 +02:00
#
# Ansible is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
2013-10-14 00:07:27 +02:00
#
# You should have received a copy of the GNU General Public License
# along with Ansible. If not, see <http://www.gnu.org/licenses/>.
######################################################################
2013-10-14 00:07:27 +02:00
DOCUMENTATION = '''
---
module: docker
version_added: "1.4"
2013-10-14 00:07:27 +02:00
short_description: manage docker containers
description:
2015-03-07 06:51:51 +01:00
- Manage the life cycle of docker containers.
2013-10-14 00:07:27 +02:00
options:
count:
description:
2015-03-07 06:51:51 +01:00
- Number of matching containers that should be in the desired state.
2013-10-14 00:07:27 +02:00
default: 1
image:
description:
2015-03-07 06:51:51 +01:00
- Container image used to match and launch containers.
2013-10-14 00:07:27 +02:00
required: true
2014-12-12 17:01:08 +01:00
pull:
description:
2015-03-07 06:51:51 +01:00
- Control when container images are updated from the C(docker_url) registry.
If "missing," images will be pulled only when missing from the host;
if '"always," the registry will be checked for a newer version of the
image' each time the task executes.
2014-12-12 17:01:08 +01:00
default: missing
choices: [ "missing", "always" ]
version_added: "1.9"
entrypoint:
description:
- Corresponds to ``--entrypoint`` option of ``docker run`` command and
``ENTRYPOINT`` directive of Dockerfile.
Used to match and launch containers.
default: null
required: false
2015-12-19 06:28:01 +01:00
version_added: "2.1"
2013-10-14 00:07:27 +02:00
command:
description:
2015-03-07 06:51:51 +01:00
- Command used to match and launch containers.
2013-10-14 00:07:27 +02:00
default: null
name:
description:
2015-03-07 06:51:51 +01:00
- Name used to match and uniquely name launched containers. Explicit names
are used to uniquely identify a single container or to link among
containers. Mutually exclusive with a "count" other than "1".
default: null
version_added: "1.5"
2013-10-14 00:07:27 +02:00
ports:
description:
2015-06-17 18:39:14 +02:00
- "List containing private to public port mapping specification.
Use docker 'CLI-style syntax: C(8000), C(9000:8000), or C(0.0.0.0:9000:8000)'
where 8000 is a container port, 9000 is a host port, and 0.0.0.0 is - a host interface.
The container ports need to be exposed either in the Dockerfile or via the C(expose) option."
2013-10-14 00:07:27 +02:00
default: null
version_added: "1.5"
expose:
description:
2015-03-07 06:51:51 +01:00
- List of additional container ports to expose for port mappings or links.
If the port is already exposed using EXPOSE in a Dockerfile, you don't
need to expose it again.
default: null
version_added: "1.5"
publish_all_ports:
description:
2015-03-07 06:51:51 +01:00
- Publish all exposed ports to the host interfaces.
default: false
version_added: "1.5"
2013-10-14 00:07:27 +02:00
volumes:
description:
- List of volumes to mount within the container
- 'Use docker CLI-style syntax: C(/host:/container[:mode])'
- You can specify a read mode for the mount with either C(ro) or C(rw).
Starting at version 2.1, SELinux hosts can additionally use C(z) or C(Z)
mount options to use a shared or private label for the volume.
2013-10-14 00:07:27 +02:00
default: null
volumes_from:
description:
2015-03-07 06:51:51 +01:00
- List of names of containers to mount volumes from.
2013-10-14 00:07:27 +02:00
default: null
links:
description:
2015-03-07 06:51:51 +01:00
- List of other containers to link within this container with an optional
- 'alias. Use docker CLI-style syntax: C(redis:myredis).'
default: null
version_added: "1.5"
devices:
description:
- List of host devices to expose to container
default: null
required: false
2015-12-19 01:33:48 +01:00
version_added: "2.1"
log_driver:
description:
- You can specify a different logging driver for the container than for the daemon.
"json-file" Default logging driver for Docker. Writes JSON messages to file.
docker logs command is available only for this logging driver.
"none" disables any logging for the container.
"syslog" Syslog logging driver for Docker. Writes log messages to syslog.
docker logs command is not available for this logging driver.
"journald" Journald logging driver for Docker. Writes log messages to "journald".
"gelf" Graylog Extended Log Format (GELF) logging driver for Docker. Writes log messages to a GELF endpoint likeGraylog or Logstash.
"fluentd" Fluentd logging driver for Docker. Writes log messages to "fluentd" (forward input).
2016-01-18 15:16:38 +01:00
"awslogs" (added in 2.1) Awslogs logging driver for Docker. Writes log messages to AWS Cloudwatch Logs.
If not defined explicitly, the Docker daemon's default ("json-file") will apply.
Requires docker >= 1.6.0.
required: false
default: json-file
choices:
- json-file
- none
- syslog
- journald
- gelf
- fluentd
- awslogs
version_added: "2.0"
2015-09-08 17:19:26 +02:00
log_opt:
description:
- Additional options to pass to the logging driver selected above. See Docker `log-driver
<https://docs.docker.com/reference/logging/overview/>` documentation for more information.
2015-09-08 17:19:26 +02:00
Requires docker >=1.7.0.
required: false
default: null
version_added: "2.0"
2013-10-14 00:07:27 +02:00
memory_limit:
description:
2015-03-07 06:51:51 +01:00
- RAM allocated to the container as a number of bytes or as a human-readable
string like "512MB". Leave as "0" to specify no limit.
2014-12-12 17:01:08 +01:00
default: 0
2013-10-14 00:07:27 +02:00
docker_url:
description:
2015-03-07 06:51:51 +01:00
- URL of the host running the docker daemon. This will default to the env
var DOCKER_HOST if unspecified.
2014-12-12 17:01:08 +01:00
default: ${DOCKER_HOST} or unix://var/run/docker.sock
use_tls:
2014-12-12 17:01:08 +01:00
description:
- Whether to use tls to connect to the docker server. "no" means not to
use tls (and ignore any other tls related parameters). "encrypt" means
to use tls to encrypt the connection to the server. "verify" means to
also verify that the server's certificate is valid for the server
(this both verifies the certificate against the CA and that the
certificate was issued for that host. If this is unspecified, tls will
only be used if one of the other tls options require it.
choices: [ "no", "encrypt", "verify" ]
version_added: "1.9"
tls_client_cert:
description:
- Path to the PEM-encoded certificate used to authenticate docker client.
If specified tls_client_key must be valid
2014-12-12 17:01:08 +01:00
default: ${DOCKER_CERT_PATH}/cert.pem
version_added: "1.9"
tls_client_key:
2014-12-12 17:01:08 +01:00
description:
- Path to the PEM-encoded key used to authenticate docker client. If
specified tls_client_cert must be valid
2014-12-12 17:01:08 +01:00
default: ${DOCKER_CERT_PATH}/key.pem
version_added: "1.9"
tls_ca_cert:
2014-12-12 17:01:08 +01:00
description:
2015-03-07 06:51:51 +01:00
- Path to a PEM-encoded certificate authority to secure the Docker connection.
This has no effect if use_tls is encrypt.
2014-12-12 17:01:08 +01:00
default: ${DOCKER_CERT_PATH}/ca.pem
version_added: "1.9"
tls_hostname:
description:
- A hostname to check matches what's supplied in the docker server's
certificate. If unspecified, the hostname is taken from the docker_url.
default: Taken from docker_url
version_added: "1.9"
docker_api_version:
description:
2015-03-07 06:51:51 +01:00
- Remote API version to use. This defaults to the current default as
specified by docker-py.
default: docker-py default remote API version
version_added: "1.8"
docker_user:
description:
- Username or UID to use within the container
required: false
2015-07-21 03:18:49 +02:00
default: null
version_added: "2.0"
2013-10-14 00:07:27 +02:00
username:
description:
2015-03-07 06:51:51 +01:00
- Remote API username.
2013-10-14 00:07:27 +02:00
default: null
password:
description:
2015-03-07 06:51:51 +01:00
- Remote API password.
2013-10-14 00:07:27 +02:00
default: null
email:
description:
2015-03-07 06:51:51 +01:00
- Remote API email.
default: null
2013-10-14 00:07:27 +02:00
hostname:
description:
2015-03-07 06:51:51 +01:00
- Container hostname.
2013-10-14 00:07:27 +02:00
default: null
domainname:
description:
2015-03-07 06:51:51 +01:00
- Container domain name.
default: null
2013-10-14 00:07:27 +02:00
env:
description:
2015-03-07 06:51:51 +01:00
- Pass a dict of environment variables to the container.
2013-10-14 00:07:27 +02:00
default: null
dns:
description:
2015-03-07 06:51:51 +01:00
- List of custom DNS servers for the container.
2013-10-14 00:07:27 +02:00
required: false
default: null
detach:
description:
2015-03-30 21:52:17 +02:00
- Enable detached mode to leave the container running in background. If
disabled, fail unless the process exits cleanly.
2013-10-14 00:07:27 +02:00
default: true
signal:
version_added: "2.0"
description:
- With the state "killed", you can alter the signal sent to the
container.
required: false
default: KILL
2013-10-14 00:07:27 +02:00
state:
description:
2015-03-07 06:51:51 +01:00
- Assert the container's desired state. "present" only asserts that the
matching containers exist. "started" asserts that the matching
containers both exist and are running, but takes no action if any
configuration has changed. "reloaded" (added in Ansible 1.9) asserts that all matching
2015-03-07 06:51:51 +01:00
containers are running and restarts any that have any images or
configuration out of date. "restarted" unconditionally restarts (or
starts) the matching containers. "stopped" and '"killed" stop and kill
all matching containers. "absent" stops and then' removes any matching
containers.
2013-10-14 00:07:27 +02:00
required: false
default: started
2014-12-12 17:01:08 +01:00
choices:
2015-03-07 06:51:51 +01:00
- present
- started
- reloaded
- restarted
- stopped
- killed
- absent
2013-10-14 00:07:27 +02:00
privileged:
description:
2015-03-07 06:51:51 +01:00
- Whether the container should run in privileged mode or not.
2013-10-14 00:07:27 +02:00
default: false
lxc_conf:
description:
2015-03-07 06:51:51 +01:00
- LXC configuration parameters, such as C(lxc.aa_profile:unconfined).
default: null
stdin_open:
description:
2015-03-07 06:51:51 +01:00
- Keep stdin open after a container is launched.
default: false
version_added: "1.6"
tty:
description:
2015-03-07 06:51:51 +01:00
- Allocate a pseudo-tty within the container.
default: false
version_added: "1.6"
net:
description:
2015-03-07 06:51:51 +01:00
- 'Network mode for the launched container: bridge, none, container:<name|id>'
- or host. Requires docker >= 0.11.
default: false
version_added: "1.8"
pid:
description:
- Set the PID namespace mode for the container (currently only supports 'host'). Requires docker-py >= 1.0.0 and docker >= 1.5.0
required: false
2015-03-06 10:13:32 +01:00
default: None
aliases: []
2015-03-06 10:13:32 +01:00
version_added: "1.9"
registry:
description:
2015-03-07 06:51:51 +01:00
- Remote registry URL to pull images from.
2014-12-12 17:01:08 +01:00
default: DockerHub
aliases: []
version_added: "1.8"
read_only:
description:
- Mount the container's root filesystem as read only
default: null
aliases: []
version_added: "2.0"
restart_policy:
description:
2015-03-07 06:51:51 +01:00
- Container restart policy.
- The 'unless-stopped' choice is only available starting in Ansible 2.1 and for Docker 1.9 and above.
2015-11-18 10:02:51 +01:00
choices: ["no", "on-failure", "always", "unless-stopped"]
2014-12-12 17:01:08 +01:00
default: null
version_added: "1.9"
restart_policy_retry:
description:
2015-03-07 06:51:51 +01:00
- Maximum number of times to restart a container. Leave as "0" for unlimited
retries.
2014-12-12 17:01:08 +01:00
default: 0
version_added: "1.9"
extra_hosts:
version_added: "2.0"
description:
- Dict of custom host-to-IP mappings to be defined in the container
insecure_registry:
description:
2015-03-07 06:51:51 +01:00
- Use insecure private registry by HTTP instead of HTTPS. Needed for
docker-py >= 0.5.0.
default: false
version_added: "1.9"
2014-11-24 23:08:12 +01:00
cpu_set:
description:
- CPUs in which to allow execution. Requires docker-py >= 0.6.0.
required: false
default: null
2015-07-21 02:09:36 +02:00
version_added: "2.0"
cap_add:
description:
- Add capabilities for the container. Requires docker-py >= 0.5.0.
required: false
default: false
version_added: "2.0"
cap_drop:
description:
- Drop capabilities for the container. Requires docker-py >= 0.5.0.
required: false
default: false
aliases: []
version_added: "2.0"
2015-09-25 22:58:20 +02:00
labels:
description:
- Set container labels. Requires docker >= 1.6 and docker-py >= 1.2.0.
required: false
2015-09-25 22:58:20 +02:00
default: null
version_added: "2.1"
stop_timeout:
description:
- How many seconds to wait for the container to stop before killing it.
required: false
default: 10
version_added: "2.0"
2016-01-09 19:01:02 +01:00
timeout:
description:
- Docker daemon response timeout in seconds.
required: false
default: 60
version_added: "2.1"
cpu_shares:
description:
- CPU shares (relative weight). Requires docker-py >= 0.6.0.
required: false
default: 0
2016-01-12 14:27:33 +01:00
version_added: "2.1"
ulimits:
description:
- ulimits, list ulimits with name, soft and optionally
hard limit separated by colons. e.g. nofile:1024:2048
Requires docker-py >= 1.2.0 and docker >= 1.6.0
required: false
default: null
version_added: "2.1"
author:
- "Cove Schneider (@cove)"
- "Joshua Conner (@joshuaconner)"
- "Pavel Antonov (@softzilla)"
- "Ash Wilson (@smashwilson)"
- "Thomas Steinbach (@ThomasSteinbach)"
- "Philippe Jandot (@zfil)"
- "Daan Oosterveld (@dusdanig)"
requirements:
- "python >= 2.6"
- "docker-py >= 0.3.0"
- "The docker server >= 0.10.0"
2013-10-14 00:07:27 +02:00
'''
2013-10-14 05:34:14 +02:00
EXAMPLES = '''
2014-12-12 17:01:08 +01:00
# Containers are matched either by name (if provided) or by an exact match of
# the image they were launched with and the command they're running. The module
# can accept either a name to target a container uniquely, or a count to operate
# on multiple containers at once when it makes sense to do so.
# Ensure that a data container with the name "mydata" exists. If no container
# by this name exists, it will be created, but not started.
- name: data container
docker:
name: mydata
image: busybox
state: present
volumes:
- /data
# Ensure that a Redis server is running, using the volume from the data
# container. Expose the default Redis port.
- name: redis container
docker:
name: myredis
image: redis
command: redis-server --appendonly yes
state: started
expose:
- 6379
volumes_from:
- mydata
# Ensure that a container of your application server is running. This will:
# - pull the latest version of your application image from DockerHub.
# - ensure that a container is running with the specified name and exact image.
# If any configuration options have changed, the existing container will be
# stopped and removed, and a new one will be launched in its place.
# - link this container to the existing redis container launched above with
# an alias.
2015-12-19 01:33:48 +01:00
# - grant the container read write permissions for the host's /dev/sda device
# through a node named /dev/xvda
2014-12-12 17:01:08 +01:00
# - bind TCP port 9000 within the container to port 8080 on all interfaces
# on the host.
# - bind UDP port 9001 within the container to port 8081 on the host, only
# listening on localhost.
# - set the environment variable SECRET_KEY to "ssssh".
- name: application container
docker:
name: myapplication
image: someuser/appimage
state: reloaded
pull: always
links:
- "myredis:aliasedredis"
devices:
- "/dev/sda:/dev/xvda:rwm"
2014-12-12 17:01:08 +01:00
ports:
- "8080:9000"
- "127.0.0.1:8081:9001/udp"
env:
SECRET_KEY: ssssh
# Ensure that exactly five containers of another server are running with this
# exact image and command. If fewer than five are running, more will be launched;
# if more are running, the excess will be stopped.
- name: load-balanced containers
docker:
state: reloaded
count: 5
image: someuser/anotherappimage
command: sleep 1d
# Unconditionally restart a service container. This may be useful within a
# handler, for example.
- name: application service
docker:
name: myservice
image: someuser/serviceimage
state: restarted
# Stop all containers running the specified image.
- name: obsolete container
docker:
image: someuser/oldandbusted
state: stopped
# Stop and remove a container with the specified name.
- name: obsolete container
docker:
name: ohno
image: someuser/oldandbusted
state: absent
2015-09-08 17:19:26 +02:00
# Example Syslogging Output
- name: myservice container
docker:
name: myservice
image: someservice/someimage
state: reloaded
log_driver: syslog
log_opt:
syslog-address: tcp://my-syslog-server:514
syslog-facility: daemon
syslog-tag: myservice
2013-10-14 05:34:14 +02:00
'''
HAS_DOCKER_PY = True
DEFAULT_DOCKER_API_VERSION = None
2016-01-09 19:01:02 +01:00
DEFAULT_TIMEOUT_SECONDS = 60
import sys
import json
import os
import shlex
from urlparse import urlparse
2013-10-14 00:07:27 +02:00
try:
import docker.client
import docker.utils
import docker.errors
from requests.exceptions import RequestException
except ImportError:
HAS_DOCKER_PY = False
if HAS_DOCKER_PY:
try:
from docker.errors import APIError as DockerAPIError
except ImportError:
from docker.client import APIError as DockerAPIError
try:
# docker-py 1.2+
import docker.constants
DEFAULT_DOCKER_API_VERSION = docker.constants.DEFAULT_DOCKER_API_VERSION
2016-01-09 19:01:02 +01:00
DEFAULT_TIMEOUT_SECONDS = docker.constants.DEFAULT_TIMEOUT_SECONDS
except (ImportError, AttributeError):
# docker-py less than 1.2
DEFAULT_DOCKER_API_VERSION = docker.client.DEFAULT_DOCKER_API_VERSION
2016-01-09 19:01:02 +01:00
DEFAULT_TIMEOUT_SECONDS = docker.client.DEFAULT_TIMEOUT_SECONDS
2013-10-14 00:07:27 +02:00
def _human_to_bytes(number):
suffixes = ['B', 'KB', 'MB', 'GB', 'TB', 'PB']
if isinstance(number, int):
return number
if number[-1] == suffixes[0] and number[-2].isdigit():
return number[:-1]
i = 1
for each in suffixes[1:]:
if number[-len(each):] == suffixes[i]:
return int(number[:-len(each)]) * (1024 ** i)
i = i + 1
raise ValueError('Could not convert %s to integer' % (number,))
2013-10-14 00:07:27 +02:00
2013-10-14 00:07:27 +02:00
def _ansible_facts(container_list):
return {"docker_containers": container_list}
2013-10-14 00:07:27 +02:00
def _docker_id_quirk(inspect):
# XXX: some quirk in docker
if 'ID' in inspect:
inspect['Id'] = inspect['ID']
del inspect['ID']
return inspect
def get_split_image_tag(image):
# If image contains a host or org name, omit that from our check
if '/' in image:
registry, resource = image.rsplit('/', 1)
else:
registry, resource = None, image
# now we can determine if image has a tag
if ':' in resource:
resource, tag = resource.split(':', 1)
if registry:
resource = '/'.join((registry, resource))
if tag == "":
tag = "latest"
else:
tag = "latest"
resource = image
return resource, tag
def normalize_image(image):
"""
Normalize a Docker image name to include the implied :latest tag.
"""
return ":".join(get_split_image_tag(image))
def is_running(container):
'''Return True if an inspected container is in a state we consider "running."'''
return container['State']['Running'] == True and not container['State'].get('Ghost', False)
def get_docker_py_versioninfo():
if hasattr(docker, '__version__'):
# a '__version__' attribute was added to the module but not until
# after 0.3.0 was pushed to pypi. If it's there, use it.
version = []
for part in docker.__version__.split('.'):
try:
version.append(int(part))
except ValueError:
for idx, char in enumerate(part):
if not char.isdigit():
nondigit = part[idx:]
digit = part[:idx]
break
if digit:
version.append(int(digit))
if nondigit:
version.append(nondigit)
elif hasattr(docker.Client, '_get_raw_response_socket'):
# HACK: if '__version__' isn't there, we check for the existence of
# `_get_raw_response_socket` in the docker.Client class, which was
# added in 0.3.0
version = (0, 3, 0)
else:
# This is untrue but this module does not function with a version less
# than 0.3.0 so it's okay to lie here.
version = (0,)
2014-12-24 03:04:04 +01:00
return tuple(version)
def check_dependencies(module):
"""
Ensure `docker-py` >= 0.3.0 is installed, and call module.fail_json with a
helpful error message if it isn't.
"""
if not HAS_DOCKER_PY:
module.fail_json(msg="`docker-py` doesn't seem to be installed, but is required for the Ansible Docker module.")
else:
versioninfo = get_docker_py_versioninfo()
if versioninfo < (0, 3, 0):
module.fail_json(msg="The Ansible Docker module requires `docker-py` >= 0.3.0.")
2014-12-18 22:43:55 +01:00
class DockerManager(object):
counters = dict(
created=0, started=0, stopped=0, killed=0, removed=0, restarted=0, pulled=0
)
reload_reasons = []
_capabilities = set()
# Map optional parameters to minimum (docker-py version, server APIVersion)
# docker-py version is a tuple of ints because we have to compare them
# server APIVersion is passed to a docker-py function that takes strings
_cap_ver_req = {
'devices': ((0, 7, 0), '1.2'),
'dns': ((0, 3, 0), '1.10'),
'volumes_from': ((0, 3, 0), '1.10'),
'restart_policy': ((0, 5, 0), '1.14'),
'extra_hosts': ((0, 7, 0), '1.3.1'),
'pid': ((1, 0, 0), '1.17'),
'log_driver': ((1, 2, 0), '1.18'),
2015-09-08 17:19:26 +02:00
'log_opt': ((1, 2, 0), '1.18'),
'host_config': ((0, 7, 0), '1.15'),
'cpu_set': ((0, 6, 0), '1.14'),
'cap_add': ((0, 5, 0), '1.14'),
'cap_drop': ((0, 5, 0), '1.14'),
'read_only': ((1, 0, 0), '1.17'),
2015-09-25 22:58:20 +02:00
'labels': ((1, 2, 0), '1.18'),
'stop_timeout': ((0, 5, 0), '1.0'),
'ulimits': ((1, 2, 0), '1.18'),
# Clientside only
'insecure_registry': ((0, 5, 0), '0.0')
}
2013-10-14 00:07:27 +02:00
def __init__(self, module):
self.module = module
self.binds = None
self.volumes = None
if self.module.params.get('volumes'):
self.binds = {}
self.volumes = []
vols = self.module.params.get('volumes')
for vol in vols:
parts = vol.split(":")
# regular volume
if len(parts) == 1:
self.volumes.append(parts[0])
# host mount (e.g. /mnt:/tmp, bind mounts host's /tmp to /mnt in the container)
elif 2 <= len(parts) <= 3:
# default to read-write
mode = 'rw'
# with supplied bind mode
if len(parts) == 3:
if parts[2] not in ["rw", "rw,Z", "rw,z", "z,rw", "Z,rw", "Z", "z", "ro", "ro,Z", "ro,z", "z,ro", "Z,ro"]:
self.module.fail_json(msg='invalid bind mode ' + parts[2])
else:
mode = parts[2]
self.binds[parts[0]] = {'bind': parts[1], 'mode': mode }
else:
self.module.fail_json(msg='volumes support 1 to 3 arguments')
self.lxc_conf = None
if self.module.params.get('lxc_conf'):
self.lxc_conf = []
options = self.module.params.get('lxc_conf')
for option in options:
parts = option.split(':', 1)
self.lxc_conf.append({"Key": parts[0], "Value": parts[1]})
self.exposed_ports = None
if self.module.params.get('expose'):
self.exposed_ports = self.get_exposed_ports(self.module.params.get('expose'))
self.port_bindings = None
if self.module.params.get('ports'):
self.port_bindings = self.get_port_bindings(self.module.params.get('ports'))
self.links = None
if self.module.params.get('links'):
self.links = self.get_links(self.module.params.get('links'))
self.env = self.module.params.get('env', None)
self.ulimits = None
if self.module.params.get('ulimits'):
self.ulimits = []
ulimits = self.module.params.get('ulimits')
for ulimit in ulimits:
parts = ulimit.split(":")
if len(parts) == 2:
self.ulimits.append({'name': parts[0], 'soft': int(parts[1]), 'hard': int(parts[1])})
elif len(parts) == 3:
self.ulimits.append({'name': parts[0], 'soft': int(parts[1]), 'hard': int(parts[2])})
else:
self.module.fail_json(msg='ulimits support 2 to 3 arguments')
2015-02-18 16:29:53 +01:00
# Connect to the docker server using any configured host and TLS settings.
env_host = os.getenv('DOCKER_HOST')
env_docker_verify = os.getenv('DOCKER_TLS_VERIFY')
2015-02-18 16:29:53 +01:00
env_cert_path = os.getenv('DOCKER_CERT_PATH')
env_docker_hostname = os.getenv('DOCKER_TLS_HOSTNAME')
2015-02-18 16:29:53 +01:00
docker_url = module.params.get('docker_url')
if not docker_url:
if env_host:
docker_url = env_host
else:
docker_url = 'unix://var/run/docker.sock'
docker_api_version = module.params.get('docker_api_version')
2016-01-09 19:01:02 +01:00
timeout = module.params.get('timeout')
2015-02-18 16:29:53 +01:00
tls_client_cert = module.params.get('tls_client_cert', None)
if not tls_client_cert and env_cert_path:
tls_client_cert = os.path.join(env_cert_path, 'cert.pem')
tls_client_key = module.params.get('tls_client_key', None)
if not tls_client_key and env_cert_path:
tls_client_key = os.path.join(env_cert_path, 'key.pem')
tls_ca_cert = module.params.get('tls_ca_cert')
if not tls_ca_cert and env_cert_path:
tls_ca_cert = os.path.join(env_cert_path, 'ca.pem')
tls_hostname = module.params.get('tls_hostname')
if tls_hostname is None:
if env_docker_hostname:
tls_hostname = env_docker_hostname
else:
parsed_url = urlparse(docker_url)
if ':' in parsed_url.netloc:
tls_hostname = parsed_url.netloc[:parsed_url.netloc.rindex(':')]
else:
tls_hostname = parsed_url
if not tls_hostname:
tls_hostname = True
# use_tls can be one of four values:
# no: Do not use tls
# encrypt: Use tls. We may do client auth. We will not verify the server
# verify: Use tls. We may do client auth. We will verify the server
# None: Only use tls if the parameters for client auth were specified
# or tls_ca_cert (which requests verifying the server with
# a specific ca certificate)
use_tls = module.params.get('use_tls')
if use_tls is None and env_docker_verify is not None:
use_tls = 'verify'
tls_config = None
if use_tls != 'no':
params = {}
# Setup client auth
if tls_client_cert and tls_client_key:
params['client_cert'] = (tls_client_cert, tls_client_key)
# We're allowed to verify the connection to the server
if use_tls == 'verify' or (use_tls is None and tls_ca_cert):
if tls_ca_cert:
params['ca_cert'] = tls_ca_cert
params['verify'] = True
params['assert_hostname'] = tls_hostname
else:
params['verify'] = True
params['assert_hostname'] = tls_hostname
elif use_tls == 'encrypt':
params['verify'] = False
2015-02-18 16:29:53 +01:00
if params:
# See https://github.com/docker/docker-py/blob/d39da11/docker/utils/utils.py#L279-L296
docker_url = docker_url.replace('tcp://', 'https://')
tls_config = docker.tls.TLSConfig(**params)
2015-02-18 16:29:53 +01:00
self.client = docker.Client(base_url=docker_url,
version=docker_api_version,
2016-01-09 19:01:02 +01:00
tls=tls_config,
timeout=timeout)
self.docker_py_versioninfo = get_docker_py_versioninfo()
2015-05-04 13:54:03 +02:00
def _check_capabilities(self):
"""
Create a list of available capabilities
"""
api_version = self.client.version()['ApiVersion']
for cap, req_vers in self._cap_ver_req.items():
if (self.docker_py_versioninfo >= req_vers[0] and
docker.utils.compare_version(req_vers[1], api_version) >= 0):
self._capabilities.add(cap)
def ensure_capability(self, capability, fail=True):
"""
Some of the functionality this ansible module implements are only
available in newer versions of docker. Ensure that the capability
is available here.
If fail is set to False then return True or False depending on whether
we have the capability. Otherwise, simply fail and exit the module if
we lack the capability.
"""
if not self._capabilities:
2015-05-04 13:54:03 +02:00
self._check_capabilities()
if capability in self._capabilities:
return True
if not fail:
return False
api_version = self.client.version()['ApiVersion']
self.module.fail_json(msg='Specifying the `%s` parameter requires'
' docker-py: %s, docker server apiversion %s; found'
' docker-py: %s, server: %s' % (
capability,
'.'.join(map(str, self._cap_ver_req[capability][0])),
self._cap_ver_req[capability][1],
'.'.join(map(str, self.docker_py_versioninfo)),
api_version))
def get_links(self, links):
"""
Parse the links passed, if a link is specified without an alias then just create the alias of the same name as the link
"""
processed_links = {}
for link in links:
parsed_link = link.split(':', 1)
if(len(parsed_link) == 2):
processed_links[parsed_link[0]] = parsed_link[1]
else:
processed_links[parsed_link[0]] = parsed_link[0]
return processed_links
def get_exposed_ports(self, expose_list):
"""
Parse the ports and protocols (TCP/UDP) to expose in the docker-py `create_container` call from the docker CLI-style syntax.
"""
if expose_list:
exposed = []
for port in expose_list:
port = str(port).strip()
if port.endswith('/tcp') or port.endswith('/udp'):
port_with_proto = tuple(port.split('/'))
else:
# assume tcp protocol if not specified
port_with_proto = (port, 'tcp')
exposed.append(port_with_proto)
return exposed
else:
return None
def get_start_params(self):
"""
Create start params
"""
params = {
'lxc_conf': self.lxc_conf,
'binds': self.binds,
'port_bindings': self.port_bindings,
'publish_all_ports': self.module.params.get('publish_all_ports'),
'privileged': self.module.params.get('privileged'),
'links': self.links,
'network_mode': self.module.params.get('net'),
}
optionals = {}
2015-12-19 01:33:48 +01:00
for optional_param in ('devices', 'dns', 'volumes_from',
'restart_policy', 'restart_policy_retry', 'pid', 'extra_hosts',
'log_driver', 'cap_add', 'cap_drop', 'read_only', 'log_opt'):
optionals[optional_param] = self.module.params.get(optional_param)
if optionals['devices'] is not None:
self.ensure_capability('devices')
params['devices'] = optionals['devices']
if optionals['dns'] is not None:
self.ensure_capability('dns')
params['dns'] = optionals['dns']
if optionals['volumes_from'] is not None:
self.ensure_capability('volumes_from')
params['volumes_from'] = optionals['volumes_from']
if optionals['restart_policy'] is not None:
self.ensure_capability('restart_policy')
params['restart_policy'] = { 'Name': optionals['restart_policy'] }
if params['restart_policy']['Name'] == 'on-failure':
params['restart_policy']['MaximumRetryCount'] = optionals['restart_policy_retry']
# docker_py only accepts 'host' or None
if 'pid' in optionals and not optionals['pid']:
optionals['pid'] = None
if optionals['pid'] is not None:
self.ensure_capability('pid')
params['pid_mode'] = optionals['pid']
if optionals['extra_hosts'] is not None:
self.ensure_capability('extra_hosts')
params['extra_hosts'] = optionals['extra_hosts']
if optionals['log_driver'] is not None:
self.ensure_capability('log_driver')
log_config = docker.utils.LogConfig(type=docker.utils.LogConfig.types.JSON)
2015-09-08 17:19:26 +02:00
if optionals['log_opt'] is not None:
for k, v in optionals['log_opt'].iteritems():
log_config.set_config_value(k, v)
log_config.type = optionals['log_driver']
params['log_config'] = log_config
if optionals['cap_add'] is not None:
self.ensure_capability('cap_add')
params['cap_add'] = optionals['cap_add']
if optionals['cap_drop'] is not None:
self.ensure_capability('cap_drop')
params['cap_drop'] = optionals['cap_drop']
if optionals['read_only'] is not None:
self.ensure_capability('read_only')
params['read_only'] = optionals['read_only']
return params
def create_host_config(self):
"""
Create HostConfig object
"""
params = self.get_start_params()
return docker.utils.create_host_config(**params)
def get_port_bindings(self, ports):
"""
Parse the `ports` string into a port bindings dict for the `start_container` call.
"""
binds = {}
for port in ports:
# ports could potentially be an array like [80, 443], so we make sure they're strings
# before splitting
parts = str(port).split(':')
container_port = parts[-1]
if '/' not in container_port:
container_port = int(parts[-1])
p_len = len(parts)
if p_len == 1:
# Bind `container_port` of the container to a dynamically
# allocated TCP port on all available interfaces of the host
# machine.
bind = ('0.0.0.0',)
elif p_len == 2:
# Bind `container_port` of the container to port `parts[0]` on
# all available interfaces of the host machine.
bind = ('0.0.0.0', int(parts[0]))
elif p_len == 3:
# Bind `container_port` of the container to port `parts[1]` on
# IP `parts[0]` of the host machine. If `parts[1]` empty bind
2015-05-04 13:54:03 +02:00
# to a dynamically allocated port of IP `parts[0]`.
bind = (parts[0], int(parts[1])) if parts[1] else (parts[0],)
if container_port in binds:
old_bind = binds[container_port]
if isinstance(old_bind, list):
# append to list if it already exists
old_bind.append(bind)
else:
# otherwise create list that contains the old and new binds
binds[container_port] = [binds[container_port], bind]
else:
binds[container_port] = bind
return binds
def get_summary_message(self):
'''
Generate a message that briefly describes the actions taken by this
task, in English.
'''
parts = []
for k, v in self.counters.iteritems():
if v == 0:
continue
if v == 1:
plural = ""
else:
plural = "s"
parts.append("%s %d container%s" % (k, v, plural))
if parts:
return ", ".join(parts) + "."
else:
return "No action taken."
def get_reload_reason_message(self):
'''
Generate a message describing why any reloaded containers were reloaded.
'''
if self.reload_reasons:
return ", ".join(self.reload_reasons)
else:
return None
2013-10-14 00:07:27 +02:00
def get_summary_counters_msg(self):
msg = ""
for k, v in self.counters.iteritems():
msg = msg + "%s %d " % (k, v)
return msg
2013-10-14 00:07:27 +02:00
def increment_counter(self, name):
self.counters[name] = self.counters[name] + 1
def has_changed(self):
for k, v in self.counters.iteritems():
if v > 0:
return True
return False
def get_inspect_image(self):
try:
return self.client.inspect_image(self.module.params.get('image'))
except DockerAPIError as e:
if e.response.status_code == 404:
return None
else:
raise e
def get_image_repo_tags(self):
image, tag = get_split_image_tag(self.module.params.get('image'))
if tag is None:
tag = 'latest'
resource = '%s:%s' % (image, tag)
for image in self.client.images(name=image):
if resource in image.get('RepoTags', []):
return image['RepoTags']
return []
def get_inspect_containers(self, containers):
inspect = []
for i in containers:
details = self.client.inspect_container(i['Id'])
details = _docker_id_quirk(details)
inspect.append(details)
return inspect
2013-10-14 00:07:27 +02:00
def get_differing_containers(self):
"""
Inspect all matching, running containers, and return those that were
started with parameters that differ from the ones that are provided
during this module run. A list containing the differing
containers will be returned, and a short string describing the specific
difference encountered in each container will be appended to
reload_reasons.
This generates the set of containers that need to be stopped and
started with new parameters with state=reloaded.
"""
running = self.get_running_containers()
current = self.get_inspect_containers(running)
defaults = self.client.info()
#Get API version
api_version = self.client.version()['ApiVersion']
image = self.get_inspect_image()
if image is None:
# The image isn't present. Assume that we're about to pull a new
# tag and *everything* will be restarted.
#
# This will give false positives if you untag an image on the host
# and there's nothing more to pull.
return current
differing = []
for container in current:
# IMAGE
# Compare the image by ID rather than name, so that containers
# will be restarted when new versions of an existing image are
# pulled.
if container['Image'] != image['Id']:
self.reload_reasons.append('image ({0} => {1})'.format(container['Image'], image['Id']))
differing.append(container)
continue
# ENTRYPOINT
expected_entrypoint = self.module.params.get('entrypoint')
if expected_entrypoint:
expected_entrypoint = shlex.split(expected_entrypoint)
actual_entrypoint = container["Config"]["Entrypoint"]
if actual_entrypoint != expected_entrypoint:
self.reload_reasons.append(
'entrypoint ({0} => {1})'
.format(actual_entrypoint, expected_entrypoint)
)
differing.append(container)
continue
# COMMAND
expected_command = self.module.params.get('command')
if expected_command:
expected_command = shlex.split(expected_command)
actual_command = container["Config"]["Cmd"]
if actual_command != expected_command:
self.reload_reasons.append('command ({0} => {1})'.format(actual_command, expected_command))
differing.append(container)
continue
# EXPOSED PORTS
expected_exposed_ports = set((image['ContainerConfig'].get('ExposedPorts') or {}).keys())
for p in (self.exposed_ports or []):
expected_exposed_ports.add("/".join(p))
actually_exposed_ports = set((container["Config"].get("ExposedPorts") or {}).keys())
if actually_exposed_ports != expected_exposed_ports:
self.reload_reasons.append('exposed_ports ({0} => {1})'.format(actually_exposed_ports, expected_exposed_ports))
differing.append(container)
continue
# VOLUMES
expected_volume_keys = set((image['ContainerConfig']['Volumes'] or {}).keys())
if self.volumes:
expected_volume_keys.update(self.volumes)
2015-03-09 16:20:14 +01:00
actual_volume_keys = set((container['Config']['Volumes'] or {}).keys())
if actual_volume_keys != expected_volume_keys:
self.reload_reasons.append('volumes ({0} => {1})'.format(actual_volume_keys, expected_volume_keys))
differing.append(container)
continue
# ULIMITS
expected_ulimit_keys = set(map(lambda x: '%s:%s:%s' % (x['name'],x['soft'],x['hard']), self.ulimits or []))
actual_ulimit_keys = set(map(lambda x: '%s:%s:%s' % (x['Name'],x['Soft'],x['Hard']), (container['HostConfig']['Ulimits'] or [])))
if actual_ulimit_keys != expected_ulimit_keys:
self.reload_reasons.append('ulimits ({0} => {1})'.format(actual_ulimit_keys, expected_ulimit_keys))
differing.append(container)
continue
# CPU_SHARES
expected_cpu_shares = self.module.params.get('cpu_shares')
actual_cpu_shares = container['HostConfig']['CpuShares']
if expected_cpu_shares and actual_cpu_shares != expected_cpu_shares:
self.reload_reasons.append('cpu_shares ({0} => {1})'.format(actual_cpu_shares, expected_cpu_shares))
differing.append(container)
continue
# MEM_LIMIT
try:
expected_mem = _human_to_bytes(self.module.params.get('memory_limit'))
except ValueError as e:
self.module.fail_json(msg=str(e))
#For v1.19 API and above use HostConfig, otherwise use Config
if docker.utils.compare_version('1.19', api_version) >= 0:
actual_mem = container['HostConfig']['Memory']
else:
actual_mem = container['Config']['Memory']
if expected_mem and actual_mem != expected_mem:
self.reload_reasons.append('memory ({0} => {1})'.format(actual_mem, expected_mem))
differing.append(container)
continue
# ENVIRONMENT
# actual_env is likely to include environment variables injected by
# the Dockerfile.
expected_env = {}
for image_env in image['ContainerConfig']['Env'] or []:
name, value = image_env.split('=', 1)
expected_env[name] = value
if self.env:
for name, value in self.env.iteritems():
expected_env[name] = str(value)
actual_env = {}
for container_env in container['Config']['Env'] or []:
name, value = container_env.split('=', 1)
actual_env[name] = value
if actual_env != expected_env:
# Don't include the environment difference in the output.
self.reload_reasons.append('environment {0} => {1}'.format(actual_env, expected_env))
differing.append(container)
continue
2015-09-25 22:58:20 +02:00
# LABELS
expected_labels = {}
for name, value in self.module.params.get('labels').iteritems():
expected_labels[name] = str(value)
actual_labels = {}
for container_label in container['Config']['Labels'] or []:
name, value = container_label.split('=', 1)
actual_labels[name] = value
if actual_labels != expected_labels:
self.reload_reasons.append('labels {0} => {1}'.format(actual_labels, expected_labels))
differing.append(container)
continue
# HOSTNAME
expected_hostname = self.module.params.get('hostname')
actual_hostname = container['Config']['Hostname']
if expected_hostname and actual_hostname != expected_hostname:
self.reload_reasons.append('hostname ({0} => {1})'.format(actual_hostname, expected_hostname))
differing.append(container)
continue
# DOMAINNAME
expected_domainname = self.module.params.get('domainname')
actual_domainname = container['Config']['Domainname']
if expected_domainname and actual_domainname != expected_domainname:
self.reload_reasons.append('domainname ({0} => {1})'.format(actual_domainname, expected_domainname))
differing.append(container)
continue
# DETACH
# We don't have to check for undetached containers. If it wasn't
# detached, it would have stopped before the playbook continued!
# NAME
# We also don't have to check name, because this is one of the
# criteria that's used to determine which container(s) match in
# the first place.
# STDIN_OPEN
expected_stdin_open = self.module.params.get('stdin_open')
actual_stdin_open = container['Config']['AttachStdin']
if actual_stdin_open != expected_stdin_open:
self.reload_reasons.append('stdin_open ({0} => {1})'.format(actual_stdin_open, expected_stdin_open))
differing.append(container)
continue
# TTY
expected_tty = self.module.params.get('tty')
actual_tty = container['Config']['Tty']
if actual_tty != expected_tty:
self.reload_reasons.append('tty ({0} => {1})'.format(actual_tty, expected_tty))
differing.append(container)
continue
# -- "start" call differences --
# LXC_CONF
if self.lxc_conf:
expected_lxc = set(self.lxc_conf)
actual_lxc = set(container['HostConfig']['LxcConf'] or [])
if actual_lxc != expected_lxc:
self.reload_reasons.append('lxc_conf ({0} => {1})'.format(actual_lxc, expected_lxc))
differing.append(container)
continue
# BINDS
expected_binds = set()
if self.binds:
for host_path, config in self.binds.iteritems():
if isinstance(config, dict):
container_path = config['bind']
mode = config['mode']
else:
container_path = config
mode = 'rw'
expected_binds.add("{0}:{1}:{2}".format(host_path, container_path, mode))
actual_binds = set()
for bind in (container['HostConfig']['Binds'] or []):
if len(bind.split(':')) == 2:
actual_binds.add(bind + ":rw")
else:
actual_binds.add(bind)
if actual_binds != expected_binds:
self.reload_reasons.append('binds ({0} => {1})'.format(actual_binds, expected_binds))
differing.append(container)
continue
# PORT BINDINGS
expected_bound_ports = {}
if self.port_bindings:
for container_port, config in self.port_bindings.iteritems():
if isinstance(container_port, int):
container_port = "{0}/tcp".format(container_port)
if len(config) == 1:
expected_bound_ports[container_port] = [{'HostIp': "0.0.0.0", 'HostPort': ""}]
elif isinstance(config[0], tuple):
expected_bound_ports[container_port] = []
for hostip, hostport in config:
expected_bound_ports[container_port].append({ 'HostIp': hostip, 'HostPort': str(hostport)})
else:
expected_bound_ports[container_port] = [{'HostIp': config[0], 'HostPort': str(config[1])}]
actual_bound_ports = container['HostConfig']['PortBindings'] or {}
if actual_bound_ports != expected_bound_ports:
self.reload_reasons.append('port bindings ({0} => {1})'.format(actual_bound_ports, expected_bound_ports))
differing.append(container)
continue
# PUBLISHING ALL PORTS
# What we really care about is the set of ports that is actually
# published. That should be caught above.
# PRIVILEGED
expected_privileged = self.module.params.get('privileged')
actual_privileged = container['HostConfig']['Privileged']
if actual_privileged != expected_privileged:
self.reload_reasons.append('privileged ({0} => {1})'.format(actual_privileged, expected_privileged))
differing.append(container)
continue
# LINKS
expected_links = set()
for link, alias in (self.links or {}).iteritems():
expected_links.add("/{0}:{1}/{2}".format(link, container["Name"], alias))
actual_links = set(container['HostConfig']['Links'] or [])
if actual_links != expected_links:
self.reload_reasons.append('links ({0} => {1})'.format(actual_links, expected_links))
differing.append(container)
continue
# NETWORK MODE
expected_netmode = self.module.params.get('net') or 'bridge'
actual_netmode = container['HostConfig']['NetworkMode'] or 'bridge'
if actual_netmode != expected_netmode:
self.reload_reasons.append('net ({0} => {1})'.format(actual_netmode, expected_netmode))
differing.append(container)
continue
2015-12-19 01:33:48 +01:00
# DEVICES
expected_devices = set()
for device in (self.module.params.get('devices') or []):
if len(device.split(':')) == 2:
expected_devices.add(device + ":rwm")
else:
expected_devices.add(device)
actual_devices = set()
for device in (container['HostConfig']['Devices'] or []):
actual_devices.add("{PathOnHost}:{PathInContainer}:{CgroupPermissions}".format(**device))
if actual_devices != expected_devices:
self.reload_reasons.append('devices ({0} => {1})'.format(actual_devices, expected_devices))
differing.append(container)
continue
# DNS
expected_dns = set(self.module.params.get('dns') or [])
actual_dns = set(container['HostConfig']['Dns'] or [])
if actual_dns != expected_dns:
self.reload_reasons.append('dns ({0} => {1})'.format(actual_dns, expected_dns))
differing.append(container)
continue
# VOLUMES_FROM
expected_volumes_from = set(self.module.params.get('volumes_from') or [])
actual_volumes_from = set(container['HostConfig']['VolumesFrom'] or [])
if actual_volumes_from != expected_volumes_from:
self.reload_reasons.append('volumes_from ({0} => {1})'.format(actual_volumes_from, expected_volumes_from))
differing.append(container)
# LOG_DRIVER
2015-09-08 17:19:26 +02:00
if self.ensure_capability('log_driver', False):
expected_log_driver = self.module.params.get('log_driver') or defaults['LoggingDriver']
actual_log_driver = container['HostConfig']['LogConfig']['Type']
if actual_log_driver != expected_log_driver:
self.reload_reasons.append('log_driver ({0} => {1})'.format(actual_log_driver, expected_log_driver))
differing.append(container)
continue
2015-09-08 17:19:26 +02:00
if self.ensure_capability('log_opt', False):
expected_logging_opts = self.module.params.get('log_opt') or {}
actual_log_opts = container['HostConfig']['LogConfig']['Config']
if len(set(expected_logging_opts.items()) - set(actual_log_opts.items())) != 0:
log_opt_reasons = {
'added': dict(set(expected_logging_opts.items()) - set(actual_log_opts.items())),
'removed': dict(set(actual_log_opts.items()) - set(expected_logging_opts.items()))
}
self.reload_reasons.append('log_opt ({0})'.format(log_opt_reasons))
differing.append(container)
return differing
2013-10-14 00:07:27 +02:00
def get_deployed_containers(self):
"""
Return any matching containers that are already present.
"""
entrypoint = self.module.params.get('entrypoint')
if entrypoint is not None:
entrypoint = shlex.split(entrypoint)
command = self.module.params.get('command')
if command is not None:
command = shlex.split(command)
name = self.module.params.get('name')
if name and not name.startswith('/'):
name = '/' + name
deployed = []
2013-10-14 00:07:27 +02:00
# "images" will be a collection of equivalent "name:tag" image names
# that map to the same Docker image.
inspected = self.get_inspect_image()
if inspected:
repo_tags = self.get_image_repo_tags()
else:
repo_tags = [normalize_image(self.module.params.get('image'))]
for container in self.client.containers(all=True):
2015-03-24 16:18:15 +01:00
details = None
if name:
name_list = container.get('Names')
if name_list is None:
name_list = []
matches = name in name_list
else:
2015-05-06 11:53:40 +02:00
details = self.client.inspect_container(container['Id'])
2015-03-24 16:18:15 +01:00
details = _docker_id_quirk(details)
running_image = normalize_image(details['Config']['Image'])
image_matches = running_image in repo_tags
command_matches = command == details['Config']['Cmd']
entrypoint_matches = (
entrypoint == details['Config']['Entrypoint']
)
matches = (image_matches and command_matches and
entrypoint_matches)
if matches:
2015-03-24 16:18:15 +01:00
if not details:
details = self.client.inspect_container(container['Id'])
2015-03-24 16:18:15 +01:00
details = _docker_id_quirk(details)
2013-10-14 00:07:27 +02:00
deployed.append(details)
return deployed
def get_running_containers(self):
return [c for c in self.get_deployed_containers() if is_running(c)]
2013-10-14 00:07:27 +02:00
def pull_image(self):
extra_params = {}
if self.module.params.get('insecure_registry'):
if self.ensure_capability('insecure_registry', fail=False):
extra_params['insecure_registry'] = self.module.params.get('insecure_registry')
resource = self.module.params.get('image')
image, tag = get_split_image_tag(resource)
if self.module.params.get('username'):
try:
self.client.login(
self.module.params.get('username'),
password=self.module.params.get('password'),
email=self.module.params.get('email'),
registry=self.module.params.get('registry')
)
2015-03-12 09:04:50 +01:00
except Exception as e:
self.module.fail_json(msg="failed to login to the remote registry, check your username/password.", error=repr(e))
try:
2015-03-15 19:22:07 +01:00
changes = list(self.client.pull(image, tag=tag, stream=True, **extra_params))
pull_success = False
for change in changes:
status = json.loads(change).get('status', '')
if status.startswith('Status: Image is up to date for'):
# Image is already up to date. Don't increment the counter.
pull_success = True
break
elif (status.startswith('Status: Downloaded newer image for') or
status.startswith('Download complete')):
# Image was updated. Increment the pull counter.
self.increment_counter('pulled')
pull_success = True
break
if not pull_success:
# Unrecognized status string.
2015-03-15 19:22:07 +01:00
self.module.fail_json(msg="Unrecognized status from pull.", status=status, changes=changes)
2015-03-12 09:04:50 +01:00
except Exception as e:
self.module.fail_json(msg="Failed to pull the specified image: %s" % resource, error=repr(e))
2013-10-14 00:07:27 +02:00
def create_containers(self, count=1):
try:
mem_limit = _human_to_bytes(self.module.params.get('memory_limit'))
except ValueError as e:
self.module.fail_json(msg=str(e))
2015-08-06 15:46:01 +02:00
api_version = self.client.version()['ApiVersion']
2013-10-14 00:07:27 +02:00
params = {'image': self.module.params.get('image'),
'entrypoint': self.module.params.get('entrypoint'),
2013-10-14 00:07:27 +02:00
'command': self.module.params.get('command'),
'ports': self.exposed_ports,
'volumes': self.volumes,
'environment': self.env,
2015-09-25 22:58:20 +02:00
'labels': self.module.params.get('labels'),
2013-10-14 00:07:27 +02:00
'hostname': self.module.params.get('hostname'),
'domainname': self.module.params.get('domainname'),
2013-10-14 00:07:27 +02:00
'detach': self.module.params.get('detach'),
'name': self.module.params.get('name'),
'stdin_open': self.module.params.get('stdin_open'),
'tty': self.module.params.get('tty'),
'cpuset': self.module.params.get('cpu_set'),
'cpu_shares': self.module.params.get('cpu_shares'),
'user': self.module.params.get('docker_user'),
2013-10-14 00:07:27 +02:00
}
if self.ensure_capability('host_config', fail=False):
params['host_config'] = self.create_host_config()
2015-08-06 15:46:01 +02:00
#For v1.19 API and above use HostConfig, otherwise use Config
if docker.utils.compare_version('1.19', api_version) < 0:
2015-08-06 15:46:01 +02:00
params['mem_limit'] = mem_limit
else:
params['host_config']['Memory'] = mem_limit
2015-08-06 15:46:01 +02:00
if self.ulimits is not None:
self.ensure_capability('ulimits')
params['host_config']['ulimits'] = self.ulimits
2015-08-06 15:46:01 +02:00
2013-10-14 00:07:27 +02:00
def do_create(count, params):
results = []
2013-10-14 06:34:58 +02:00
for _ in range(count):
2013-10-14 00:07:27 +02:00
result = self.client.create_container(**params)
self.increment_counter('created')
results.append(result)
return results
try:
containers = do_create(count, params)
except docker.errors.APIError as e:
if e.response.status_code != 404:
raise
self.pull_image()
2013-10-14 00:07:27 +02:00
containers = do_create(count, params)
return containers
def start_containers(self, containers):
params = {}
if not self.ensure_capability('host_config', fail=False):
params = self.get_start_params()
2013-10-14 00:07:27 +02:00
for i in containers:
self.client.start(i)
self.increment_counter('started')
2013-10-14 00:07:27 +02:00
2015-03-30 21:52:17 +02:00
if not self.module.params.get('detach'):
status = self.client.wait(i['Id'])
if status != 0:
output = self.client.logs(i['Id'], stdout=True, stderr=True,
stream=False, timestamps=False)
self.module.fail_json(status=status, msg=output)
2013-10-14 00:07:27 +02:00
def stop_containers(self, containers):
for i in containers:
self.client.stop(i['Id'], self.module.params.get('stop_timeout'))
2013-10-14 00:07:27 +02:00
self.increment_counter('stopped')
return [self.client.wait(i['Id']) for i in containers]
def remove_containers(self, containers):
for i in containers:
self.client.remove_container(i['Id'])
self.increment_counter('removed')
2013-10-14 00:07:27 +02:00
def kill_containers(self, containers):
for i in containers:
self.client.kill(i['Id'], self.module.params.get('signal'))
2013-10-14 00:07:27 +02:00
self.increment_counter('killed')
def restart_containers(self, containers):
for i in containers:
self.client.restart(i['Id'])
self.increment_counter('restarted')
class ContainerSet:
def __init__(self, manager):
self.manager = manager
self.running = []
self.deployed = []
self.changed = []
def refresh(self):
'''
Update our view of the matching containers from the Docker daemon.
'''
self.deployed = self.manager.get_deployed_containers()
self.running = [c for c in self.deployed if is_running(c)]
def notice_changed(self, containers):
'''
Record a collection of containers as "changed".
'''
self.changed.extend(containers)
def present(manager, containers, count, name):
'''Ensure that exactly `count` matching containers exist in any state.'''
containers.refresh()
delta = count - len(containers.deployed)
if delta > 0:
created = manager.create_containers(delta)
containers.notice_changed(manager.get_inspect_containers(created))
if delta < 0:
# If both running and stopped containers exist, remove
# stopped containers first.
containers.deployed.sort(lambda cx, cy: cmp(is_running(cx), is_running(cy)))
to_stop = []
to_remove = []
for c in containers.deployed[0:-delta]:
if is_running(c):
to_stop.append(c)
to_remove.append(c)
manager.stop_containers(to_stop)
containers.notice_changed(manager.get_inspect_containers(to_remove))
manager.remove_containers(to_remove)
def started(manager, containers, count, name):
'''Ensure that exactly `count` matching containers exist and are running.'''
containers.refresh()
delta = count - len(containers.running)
if delta > 0:
if name and containers.deployed:
# A stopped container exists with the requested name.
# Clean it up before attempting to start a new one.
manager.remove_containers(containers.deployed)
created = manager.create_containers(delta)
manager.start_containers(created)
containers.notice_changed(manager.get_inspect_containers(created))
if delta < 0:
excess = containers.running[0:-delta]
containers.notice_changed(manager.get_inspect_containers(excess))
manager.stop_containers(excess)
manager.remove_containers(excess)
def reloaded(manager, containers, count, name):
'''
Ensure that exactly `count` matching containers exist and are
running. If any associated settings have been changed (volumes,
ports or so on), restart those containers.
'''
containers.refresh()
for container in manager.get_differing_containers():
manager.stop_containers([container])
manager.remove_containers([container])
started(manager, containers, count, name)
def restarted(manager, containers, count, name):
'''
Ensure that exactly `count` matching containers exist and are
running. Unconditionally restart any that were already running.
'''
containers.refresh()
for container in manager.get_differing_containers():
manager.stop_containers([container])
manager.remove_containers([container])
containers.refresh()
manager.restart_containers(containers.running)
started(manager, containers, count, name)
def stopped(manager, containers, count, name):
'''Stop any matching containers that are running.'''
containers.refresh()
manager.stop_containers(containers.running)
containers.notice_changed(manager.get_inspect_containers(containers.running))
def killed(manager, containers, count, name):
'''Kill any matching containers that are running.'''
containers.refresh()
manager.kill_containers(containers.running)
containers.notice_changed(manager.get_inspect_containers(containers.running))
def absent(manager, containers, count, name):
'''Stop and remove any matching containers.'''
containers.refresh()
manager.stop_containers(containers.running)
containers.notice_changed(manager.get_inspect_containers(containers.deployed))
manager.remove_containers(containers.deployed)
2013-10-14 00:07:27 +02:00
def main():
module = AnsibleModule(
argument_spec = dict(
count = dict(default=1),
image = dict(required=True),
pull = dict(required=False, default='missing', choices=['missing', 'always']),
entrypoint = dict(required=False, default=None, type='str'),
2013-10-14 00:07:27 +02:00
command = dict(required=False, default=None),
expose = dict(required=False, default=None, type='list'),
ports = dict(required=False, default=None, type='list'),
publish_all_ports = dict(default=False, type='bool'),
volumes = dict(default=None, type='list'),
2013-10-14 00:07:27 +02:00
volumes_from = dict(default=None),
links = dict(default=None, type='list'),
devices = dict(default=None, type='list'),
2013-10-14 00:07:27 +02:00
memory_limit = dict(default=0),
memory_swap = dict(default=0),
cpu_shares = dict(default=0),
2015-02-18 16:29:53 +01:00
docker_url = dict(),
use_tls = dict(default=None, choices=['no', 'encrypt', 'verify']),
tls_client_cert = dict(required=False, default=None, type='str'),
tls_client_key = dict(required=False, default=None, type='str'),
tls_ca_cert = dict(required=False, default=None, type='str'),
tls_hostname = dict(required=False, type='str', default=None),
docker_api_version = dict(required=False, default=DEFAULT_DOCKER_API_VERSION, type='str'),
docker_user = dict(default=None),
username = dict(default=None),
2013-10-14 00:07:27 +02:00
password = dict(),
email = dict(),
registry = dict(),
2013-10-14 00:07:27 +02:00
hostname = dict(default=None),
domainname = dict(default=None),
env = dict(type='dict'),
2013-10-14 00:07:27 +02:00
dns = dict(),
detach = dict(default=True, type='bool'),
state = dict(default='started', choices=['present', 'started', 'reloaded', 'restarted', 'stopped', 'killed', 'absent', 'running']),
signal = dict(default=None),
2015-11-18 10:02:51 +01:00
restart_policy = dict(default=None, choices=['always', 'on-failure', 'no', 'unless-stopped']),
restart_policy_retry = dict(default=0, type='int'),
extra_hosts = dict(type='dict'),
2013-10-14 00:07:27 +02:00
debug = dict(default=False, type='bool'),
privileged = dict(default=False, type='bool'),
stdin_open = dict(default=False, type='bool'),
tty = dict(default=False, type='bool'),
lxc_conf = dict(default=None, type='list'),
name = dict(default=None),
net = dict(default=None),
pid = dict(default=None),
insecure_registry = dict(default=False, type='bool'),
log_driver = dict(default=None, choices=['json-file', 'none', 'syslog', 'journald', 'gelf', 'fluentd', 'awslogs']),
2015-09-08 17:19:26 +02:00
log_opt = dict(default=None, type='dict'),
2014-11-24 23:08:12 +01:00
cpu_set = dict(default=None),
cap_add = dict(default=None, type='list'),
cap_drop = dict(default=None, type='list'),
read_only = dict(default=None, type='bool'),
2015-09-25 22:58:20 +02:00
labels = dict(default={}, type='dict'),
stop_timeout = dict(default=10, type='int'),
2016-01-09 19:01:02 +01:00
timeout = dict(required=False, default=DEFAULT_TIMEOUT_SECONDS, type='int'),
ulimits = dict(default=None, type='list'),
),
required_together = (
['tls_client_cert', 'tls_client_key'],
),
2013-10-14 00:07:27 +02:00
)
check_dependencies(module)
2013-10-14 00:07:27 +02:00
try:
manager = DockerManager(module)
2013-10-14 00:07:27 +02:00
count = int(module.params.get('count'))
name = module.params.get('name')
pull = module.params.get('pull')
state = module.params.get('state')
if state == 'running':
# Renamed running to started in 1.9
state = 'started'
if count < 0:
module.fail_json(msg="Count must be greater than zero")
if count > 1 and name:
module.fail_json(msg="Count and name must not be used together")
2013-10-14 00:07:27 +02:00
# Explicitly pull new container images, if requested. Do this before
# noticing running and deployed containers so that the image names
# will differ if a newer image has been pulled.
# Missing images should be pulled first to avoid downtime when old
# container is stopped, but image for new one is now downloaded yet.
# It also prevents removal of running container before realizing
# that requested image cannot be retrieved.
if pull == "always" or (state == 'reloaded' and manager.get_inspect_image() is None):
manager.pull_image()
containers = ContainerSet(manager)
if state == 'present':
present(manager, containers, count, name)
elif state == 'started':
started(manager, containers, count, name)
elif state == 'reloaded':
reloaded(manager, containers, count, name)
elif state == 'restarted':
restarted(manager, containers, count, name)
elif state == 'stopped':
stopped(manager, containers, count, name)
elif state == 'killed':
killed(manager, containers, count, name)
elif state == 'absent':
absent(manager, containers, count, name)
else:
module.fail_json(msg='Unrecognized state %s. Must be one of: '
'present; started; reloaded; restarted; '
'stopped; killed; absent.' % state)
module.exit_json(changed=manager.has_changed(),
msg=manager.get_summary_message(),
summary=manager.counters,
reload_reasons=manager.get_reload_reason_message(),
ansible_facts=_ansible_facts(containers.changed))
2013-10-14 00:07:27 +02:00
except DockerAPIError as e:
module.fail_json(changed=manager.has_changed(), msg="Docker API Error: %s" % e.explanation)
2013-10-14 00:07:27 +02:00
except RequestException as e:
module.fail_json(changed=manager.has_changed(), msg=repr(e))
2013-12-02 21:13:49 +01:00
# import module snippets
from ansible.module_utils.basic import *
2013-10-14 00:07:27 +02:00
if __name__ == '__main__':
main()