diff --git a/doc/source/index.rst b/doc/source/index.rst
index 14b174838..241806f37 100644
--- a/doc/source/index.rst
+++ b/doc/source/index.rst
@@ -17,7 +17,8 @@ Index
.. toctree::
- troubleshooting
+ troubleshooting
+ metrics
How it works
============
diff --git a/doc/source/metrics.rst b/doc/source/metrics.rst
new file mode 100644
index 000000000..8d9cd3ae2
--- /dev/null
+++ b/doc/source/metrics.rst
@@ -0,0 +1,53 @@
+.. _metrics:
+
+===============================================
+Emitting metrics from Ironic-Python-Agent (IPA)
+===============================================
+
+This document describes how to emit metrics from IPA, including timers and
+counters in code to directly emitting hardware metrics from a custom
+HardwareManager.
+
+Overview
+========
+IPA uses the metrics implementation from ironic-lib, with a few caveats due
+to the dynamic configuration done at lookup time. You cannot cache the metrics
+instance as the MetricsLogger returned will change after lookup if configs
+different than the default setting have been used. This also means that the
+method decorator supported by ironic-lib cannot be used in IPA.
+
+Using a context manager
+=======================
+Using the context manager is the recommended way for sending metrics that time
+or count sections of code. However, given that you cannot cache the
+MetricsLogger, you have to explicitly call get_metrics_logger() from
+ironic-lib every time. For example:
+
+ from ironic_lib import metrics_utils
+
+ def my_method():
+ with metrics_utils.get_metrics_logger(__name__).timer():
+ return _do_work()
+
+As a note, these metric collectors do work for custom HardwareManagers as
+well, however, you may want to metric the portions of a method that determine
+compatability separate from portions of a method that actually do work, in
+order to assure the metrics are relevant and useful on all hardware.
+
+Explicitly sending metrics
+==========================
+A feature that may be particularly helpful for deployers writing custom
+HardwareManagers is the ability to explicitly send metrics. As an example,
+you could add a cleaning step which would retrieve metrics about a device and
+ship them using the provided metrics library. For example:
+
+ from ironic_lib import metrics_utils
+
+ def my_cleaning_step():
+ for name, value in _get_smart_data():
+ metrics_utils.get_metrics_logger(__name__).send_gauge(name, value)
+
+References
+==========
+For more information, please read the source of the metrics module in
+`ironic-lib <http://git.openstack.org/cgit/openstack/ironic-lib/tree/ironic_lib>`_.
diff --git a/ironic_python_agent/agent.py b/ironic_python_agent/agent.py
index 0ff3aa67e..569b2d412 100644
--- a/ironic_python_agent/agent.py
+++ b/ironic_python_agent/agent.py
@@ -20,6 +20,7 @@ import threading
import time
from oslo_concurrency import processutils
+from oslo_config import cfg
from oslo_log import log
import pkg_resources
from six.moves.urllib import parse as urlparse
@@ -35,7 +36,6 @@ from ironic_python_agent import inspector
from ironic_python_agent import ironic_api_client
from ironic_python_agent import utils
-
LOG = log.getLogger(__name__)
# Time(in seconds) to wait for any of the interfaces to be up
@@ -45,6 +45,9 @@ NETWORK_WAIT_TIMEOUT = 60
# Time(in seconds) to wait before reattempt
NETWORK_WAIT_RETRY = 5
+cfg.CONF.import_group('metrics', 'ironic_lib.metrics_utils')
+cfg.CONF.import_group('metrics_statsd', 'ironic_lib.metrics_statsd')
+
def _time():
"""Wraps time.time() for simpler testing."""
@@ -342,6 +345,15 @@ class IronicPythonAgent(base.ExecuteCommandMixin):
hardware.cache_node(self.node)
self.heartbeat_timeout = content['config']['heartbeat_timeout']
+ # Update config with values from Ironic
+ config = content.get('config', {})
+ if config.get('metrics'):
+ for opt, val in config.items():
+ setattr(cfg.CONF.metrics, opt, val)
+ if config.get('metrics_statsd'):
+ for opt, val in config.items():
+ setattr(cfg.CONF.metrics_statsd, opt, val)
+
wsgi = simple_server.make_server(
self.listen_address[0],
self.listen_address[1],
diff --git a/ironic_python_agent/api/controllers/root.py b/ironic_python_agent/api/controllers/root.py
index 46caaad01..c95e2186d 100644
--- a/ironic_python_agent/api/controllers/root.py
+++ b/ironic_python_agent/api/controllers/root.py
@@ -12,9 +12,9 @@
# License for the specific language governing permissions and limitations
# under the License.
+from ironic_lib import metrics_utils
import pecan
from pecan import rest
-
from wsme import types as wtypes
import wsmeext.pecan as wsme_pecan
@@ -81,7 +81,8 @@ class RootController(rest.RestController):
# NOTE: The reason why convert() it's being called for every
# request is because we need to get the host url from
# the request object to make the links.
- return Root.convert()
+ with metrics_utils.get_metrics_logger(__name__).timer('get'):
+ return Root.convert()
@pecan.expose()
def _route(self, args):
diff --git a/ironic_python_agent/api/controllers/v1/command.py b/ironic_python_agent/api/controllers/v1/command.py
index e4483e292..b6971fa76 100644
--- a/ironic_python_agent/api/controllers/v1/command.py
+++ b/ironic_python_agent/api/controllers/v1/command.py
@@ -13,6 +13,7 @@
# License for the specific language governing permissions and limitations
# under the License.
+from ironic_lib import metrics_utils
import pecan
from pecan import rest
from wsme import types
@@ -78,9 +79,10 @@ class CommandController(rest.RestController):
@wsme_pecan.wsexpose(CommandResultList)
def get_all(self):
"""Get all command results."""
- agent = pecan.request.agent
- results = agent.list_command_results()
- return CommandResultList.from_results(results)
+ with metrics_utils.get_metrics_logger(__name__).timer('get_all'):
+ agent = pecan.request.agent
+ results = agent.list_command_results()
+ return CommandResultList.from_results(results)
@wsme_pecan.wsexpose(CommandResult, types.text, types.text)
def get_one(self, result_id, wait=None):
@@ -91,13 +93,14 @@ class CommandController(rest.RestController):
:returns: a :class:`ironic_python_agent.api.controller.v1.command.
CommandResult` object.
"""
- agent = pecan.request.agent
- result = agent.get_command_result(result_id)
+ with metrics_utils.get_metrics_logger(__name__).timer('get_one'):
+ agent = pecan.request.agent
+ result = agent.get_command_result(result_id)
- if wait and wait.lower() == 'true':
- result.join()
+ if wait and wait.lower() == 'true':
+ result.join()
- return CommandResult.from_result(result)
+ return CommandResult.from_result(result)
@wsme_pecan.wsexpose(CommandResult, types.text, body=Command)
def post(self, wait=None, command=None):
@@ -109,14 +112,15 @@ class CommandController(rest.RestController):
:returns: a :class:`ironic_python_agent.api.controller.v1.command.
CommandResult` object.
"""
- # the POST body is always the last arg,
- # so command must be a kwarg here
- if command is None:
- command = Command()
- agent = pecan.request.agent
- result = agent.execute_command(command.name, **command.params)
+ with metrics_utils.get_metrics_logger(__name__).timer('post'):
+ # the POST body is always the last arg,
+ # so command must be a kwarg here
+ if command is None:
+ command = Command()
+ agent = pecan.request.agent
+ result = agent.execute_command(command.name, **command.params)
- if wait and wait.lower() == 'true':
- result.join()
+ if wait and wait.lower() == 'true':
+ result.join()
- return result
+ return result
diff --git a/ironic_python_agent/api/controllers/v1/status.py b/ironic_python_agent/api/controllers/v1/status.py
index 84efc6a49..b1ed83b8a 100644
--- a/ironic_python_agent/api/controllers/v1/status.py
+++ b/ironic_python_agent/api/controllers/v1/status.py
@@ -13,6 +13,7 @@
# License for the specific language governing permissions and limitations
# under the License.
+from ironic_lib import metrics_utils
import pecan
from pecan import rest
from wsme import types
@@ -48,6 +49,7 @@ class StatusController(rest.RestController):
@wsme_pecan.wsexpose(AgentStatus)
def get_all(self):
"""Get current status of the running agent."""
- agent = pecan.request.agent
- status = agent.get_status()
- return AgentStatus.from_agent_status(status)
+ with metrics_utils.get_metrics_logger(__name__).timer('get_all'):
+ agent = pecan.request.agent
+ status = agent.get_status()
+ return AgentStatus.from_agent_status(status)
diff --git a/ironic_python_agent/extensions/standby.py b/ironic_python_agent/extensions/standby.py
index 1c77d6675..a3d1ccac7 100644
--- a/ironic_python_agent/extensions/standby.py
+++ b/ironic_python_agent/extensions/standby.py
@@ -533,7 +533,6 @@ class StandbyExtension(base.BaseAgentExtension):
stream_raw_images = image_info.get('stream_raw_images', False)
# don't write image again if already cached
if self.cached_image_id != image_info['id']:
-
if self.cached_image_id is not None:
LOG.debug('Already had %s cached, overwriting',
self.cached_image_id)