Check swarm (#23)

Add swarm and service commands
Updated documentation
Added check_swarm to various test configurations

.codeclimate.yml

@@ -4,4 +4,5 @@ languages:
   PHP: true
   Python: true
 exclude_paths:
-- "test_check_docker.py"
+- "test_check_docker.py"
+- "test_check_swarm.py"

.travis.yml

@@ -8,5 +8,7 @@ install:
   - pip install codeclimate-test-reporter coverage==4.3.4 pyfakefs
 # command to run tests
 script:
-  - coverage run ./test_check_docker.py
+  - COVERAGE_FILE=.coverage.check_docker coverage run ./test_check_docker.py
+  - COVERAGE_FILE=.coverage.check_swarm coverage run ./test_check_swarm.py
+  - coverage combine .coverage.check_*
   - codeclimate-test-reporter || echo "Ignoring Code Climate reporter upload failure"

MANIFEST.in

@@ -1,4 +1,5 @@
 include license.txt
 include README.txt
 include check_docker
+include check_swarm
 include setup.py

README.md

@@ -2,24 +2,32 @@
 [![Code Climate](https://codeclimate.com/github/timdaman/check_docker/badges/gpa.svg)](https://codeclimate.com/github/timdaman/check_docker)
 [![Test Coverage](https://codeclimate.com/github/timdaman/check_docker/badges/coverage.svg)](https://codeclimate.com/github/timdaman/check_docker/coverage)
 # check_docker
-This a a nagios/NRPE compatible plugin for checking docker containers. So far you can use it to check
 
-- memory consumption in absolute units (bytes, kb, mb, gb) and as a percentage (0-100%)
-  of the container limit.
+Nagios/NRPE compatible plugins for checking docker based services. Currently there are two nagios checks
+
+- check_docker which checks docker container health
+- check_swarm which checks health of swarm nodes and services
+
+With check_docker can use it to check and alert on
+
+- memory consumption in absolute units (bytes, kb, mb, gb) and as a percentage (0-100%) of the container limit.
 - CPU usages as a percentage (0-100%) of container limit.
 - automatic restarts performed by the docker daemon
 - container status, i.e. is it running?
 - container health checks are passing?
 - uptime, i.e. is it able to stay running for a long enough time?
 - the presence of a container or containers matching specified names
-- image version (experimental!), does the running image match that in
-  the remote registry?
+- image version (experimental!), does the running image match that in the remote registry?
 
+With check_swarm you can alert
 
-This check can communicate with a local docker daemon socket file (default) or with local
+- if a node is not joined to a docker swarm
+- if a service is running in a swarm
+
+These checks can communicate with a local docker daemon socket file (default) or with local
 or remote docker daemons using secure and non-secure TCP connections.
 
-This plugin requires python 3. It is tested on 3.3 and greater but may work on older
+These plugins require python 3. It is tested on 3.3 and greater but may work on older
 versions of 3.
 
 ## Installation
@@ -33,15 +41,17 @@ With pip
 With curl
 
     curl -o /usr/local/bin/check_docker https://raw.githubusercontent.com/timdaman/check_docker/master/check_docker
-    chmod a+rx /usr/local/bin/check_docker
+    curl -o /usr/local/bin/check_swarm https://raw.githubusercontent.com/timdaman/check_docker/master/check_swarm
+    chmod a+rx /usr/local/bin/check_docker /usr/local/bin/check_swarm
 
 With wget
 
     wget -O /usr/local/bin/check_docker https://raw.githubusercontent.com/timdaman/check_docker/master/check_docker
-    chmod a+rx /usr/local/bin/check_docker
+    wget -O /usr/local/bin/check_swarm https://raw.githubusercontent.com/timdaman/check_docker/master/check_swarm
+    chmod a+rx /usr/local/bin/check_docker /usr/local/bin/check_swarm
 
 
-## Usage
+## check_docker Usage
 
     usage: check_docker [-h]
                         [--connection [/<path to>/docker.socket|<ip/host address>:<port>]
@@ -83,9 +93,35 @@ With wget
                             images. Only works with public registry.
       --restarts WARN:CRIT  Container restart thresholds.
 
-Gotchas: 
+## check_swarm Usage
 
-* When using this with older versions of docker (I have seen 1.4 and
-1.5) –status only supports ‘running’, ‘restarting’, and ‘paused’.
-* When no container is specified all containers are checked. Some containers will return critcal status because the 
-selected check(s) require a running container. 
+    usage: check_swarm [-h]
+                     [--connection [/<path to>/docker.socket|<ip/host address>:<port>]
+                     | --secure-connection [<ip/host address>:<port>]]
+                     [--timeout TIMEOUT]
+                     (--swarm | --service SERVICE [SERVICE ...])
+
+    Check docker swarm.
+
+    optional arguments:
+    -h, --help            show this help message and exit
+    --connection [/<path to>/docker.socket|<ip/host address>:<port>]
+                          Where to find docker daemon socket. (default:
+                          /var/run/docker.sock)
+    --secure-connection [<ip/host address>:<port>]
+                          Where to find TLS protected docker daemon socket.
+    --timeout TIMEOUT     Connection timeout in seconds. (default: 10.0)
+    --swarm               Check swarm status
+    --service SERVICE [SERVICE ...]
+                          One or more RegEx that match the names of the
+                          services(s) to check.
+    usage: check_swarm [-h]
+                     [--connection [/<path to>/docker.socket|<ip/host address>:<port>]
+                     | --secure-connection [<ip/host address>:<port>]]
+                     [--timeout TIMEOUT]
+                     (--swarm | --service SERVICE [SERVICE ...])
+
+Gotchas:
+
+* When using check_docker with older versions of docker (I have seen 1.4 and 1.5) –status only supports ‘running’, ‘restarting’, and ‘paused’.
+* When using check_docker, if no container is specified, all containers are checked. Some containers may return critcal status if the selected check(s) require a running container.

README.txt

@@ -5,8 +5,12 @@
 check_docker
 ============
 
-This a nagios/NRPE compatible plugin for checking docker containers. So far you
-can use it to check and alert on
+Nagios/NRPE compatible plugins for checking docker based services. Currently there are two nagios checks
+
+-  `check_docker` which checks docker container health
+-  `check_swarm` which checks health of swarm nodes and services
+
+With check_docker can use it to check and alert on
 
 -  memory consumption in absolute units (bytes, kb, mb, gb) and as a percentage (0-100%)
    of the container limit.
@@ -19,14 +23,19 @@ can use it to check and alert on
 -  image version (experimental!), does the running image match that in
    the remote registry?
 
-This check can communicate with a local docker daemon socket file (default) or with local
+With check_swarm you can alert
+
+-  if a node is not joined to a docker swarm
+-  if a service is running in a swarm
+
+These checks can communicate with a local docker daemon socket file (default) or with local
 or remote docker daemons using secure and non-secure TCP connections.
 
-This plugin requires python 3. It is tested on 3.3 and greater but may work on older
+These plugins require python 3. It is tested on 3.3 and greater but may work on older
 versions of 3.
 
-Usage
------
+check_docker Usage
+------------------
 
 ::
 
@@ -70,10 +79,41 @@ Usage
                           images. Only works with public registry.
     --restarts WARN:CRIT  Container restart thresholds.
 
+check_swarm Usage
+-----------------
+
+::
+
+  usage: check_swarm [-h]
+                     [--connection [/<path to>/docker.socket|<ip/host address>:<port>]
+                     | --secure-connection [<ip/host address>:<port>]]
+                     [--timeout TIMEOUT]
+                     (--swarm | --service SERVICE [SERVICE ...])
+
+  Check docker swarm.
+
+  optional arguments:
+    -h, --help            show this help message and exit
+    --connection [/<path to>/docker.socket|<ip/host address>:<port>]
+                          Where to find docker daemon socket. (default:
+                          /var/run/docker.sock)
+    --secure-connection [<ip/host address>:<port>]
+                          Where to find TLS protected docker daemon socket.
+    --timeout TIMEOUT     Connection timeout in seconds. (default: 10.0)
+    --swarm               Check swarm status
+    --service SERVICE [SERVICE ...]
+                          One or more RegEx that match the names of the
+                          services(s) to check.
+  usage: check_swarm [-h]
+                     [--connection [/<path to>/docker.socket|<ip/host address>:<port>]
+                     | --secure-connection [<ip/host address>:<port>]]
+                     [--timeout TIMEOUT]
+                     (--swarm | --service SERVICE [SERVICE ...])
+
 Gotchas:
 
-- When using this with older versions of docker (I have seen 1.4 and 1.5) –status only supports ‘running’, ‘restarting’, and ‘paused’.
-- When no container is specified all containers are checked. Some containers will return critcal status because the selected check(s) require a running container.
+-  When using check_docker with older versions of docker (I have seen 1.4 and 1.5) –status only supports ‘running’, ‘restarting’, and ‘paused’.
+-  When using check_docker, if no container is specified, all containers are checked. Some containers may return critcal status if the selected check(s) require a running container.
 
 .. |Build Status| image:: https://travis-ci.org/timdaman/check_docker.svg?branch=master
    :target: https://travis-ci.org/timdaman/check_docker

check_swarm

@@ -130,7 +130,7 @@ def get_services(names):
             if re.match("^{}$".format(matcher), candidate):
                 filtered.add(candidate)
                 found = True
-        # If we don't find a container that matches out regex
+        # If we don't find a service that matches out regex
         if not found:
             critical("No services match {}".format(matcher))
 
@@ -186,7 +186,7 @@ def process_url_status(status, ok_msg=None, critical_msg=None, unknown_msg=None)
 
 
 def process_args(args):
-    parser = argparse.ArgumentParser(description='Check docker containers.')
+    parser = argparse.ArgumentParser(description='Check docker swarm.')
 
     # Connect to local socket or ip address
     connection_group = parser.add_mutually_exclusive_group()

test_check_swarm.py

@@ -251,20 +251,20 @@ def test_check_service_called(self):
 
     def test_check_service_results_OK(self):
         args = ['--service', 'FOO']
-        with patch('check_swarm.get_services', return_value=(['FOO','BAR'], 200)):
+        with patch('check_swarm.get_services', return_value=['FOO','BAR']):
             with patch('check_swarm.get_service_info', return_value=(self.service, 200)):
                     check_swarm.perform_checks(args)
                     self.assertEqual(check_swarm.rc, check_swarm.OK_RC)
 
     def test_check_service_results_FAIL_missing(self):
         args = ['--service', 'missing1']
-        with patch('check_swarm.get_service_info', return_value=('', 404)):
+        with patch('check_swarm.get_url', return_value=(self.services, 200)):
                 check_swarm.perform_checks(args)
                 self.assertEqual(check_swarm.rc, check_swarm.CRITICAL_RC)
 
     def test_check_service_results_FAIL_unknown(self):
         args = ['--service', 'FOO']
-        with patch('check_swarm.get_services', return_value=(['FOO','BAR'], 200)):
+        with patch('check_swarm.get_services', return_value=['FOO','BAR']):
             with patch('check_swarm.get_service_info', return_value=('', 500)):
                 check_swarm.perform_checks(args)
                 self.assertEqual(check_swarm.rc, check_swarm.UNKNOWN_RC)