move formatter to logger file, settings reload on startup

awx/conf/apps.py

@@ -2,6 +2,8 @@
 from django.apps import AppConfig
 # from django.core import checks
 from django.utils.translation import ugettext_lazy as _
+from django.utils.log import configure_logging
+from django.conf import settings
 
 
 class ConfConfig(AppConfig):
@@ -13,4 +15,5 @@ class ConfConfig(AppConfig):
         self.module.autodiscover()
         from .settings import SettingsWrapper
         SettingsWrapper.initialize()
+        configure_logging(settings.LOGGING_CONFIG, settings.LOGGING)
         # checks.register(SettingsWrapper._check_settings)

awx/conf/signals.py

@@ -7,6 +7,7 @@ from django.core.cache import cache
 from django.core.signals import setting_changed
 from django.db.models.signals import post_save, pre_delete, post_delete
 from django.dispatch import receiver
+from django.utils.log import configure_logging
 
 # Tower
 import awx.main.signals
@@ -42,8 +43,9 @@ def handle_setting_change(key, for_delete=False):
         )
         # TODO: Move logic to task to run on all cluster nodes
         if setting_key.startswith('LOG_AGGREGATOR_'):
-            settings.LOGGING_CONFIG = None
+            configure_logging(settings.LOGGING_CONFIG, settings.LOGGING)
-            logging.config.dictConfig(settings.LOGGING)
+            # settings.LOGGING_CONFIG = None
+            # logging.config.dictConfig(settings.LOGGING)
 
 
 @receiver(post_save, sender=Setting)

awx/main/conf.py

@@ -285,3 +285,14 @@ register(
     category=_('Logging'),
     category_slug='logging',
 )
+register(
+    'LOG_AGGREGATOR_INDIVIDUAL_FACTS',
+    field_class=fields.BooleanField,
+    default=False,
+    label=_('Flag denoting to send individual messages for each fact in system tracking'),
+    help_text=_('If not set, the data from system tracking will be sent inside '
+                'of a single dictionary, but if set, separate requests will be sent '
+                'for each package, service, etc. that is found in the scan.'),
+    category=_('Logging'),
+    category_slug='logging',
+)

awx/main/consumers.py

@@ -11,6 +11,7 @@ from awx.main.models.organization import AuthToken
 
 logger = logging.getLogger('awx.main.consumers')
 
+
 def discard_groups(message):
     if 'groups' in message.channel_session:
         for group in message.channel_session['groups']:

awx/main/log_utils/formatters.py

@@ -3,50 +3,108 @@
 
 from logstash.formatter import LogstashFormatterVersion1
 from django.conf import settings
+from copy import copy
+import json
+import time
 
 
-# # Loggly example
-# 'json': {
-#     'format': '{ 
-#     "loggerName":"%(name)s", 
-#     "asciTime":"%(asctime)s", 
-#     "fileName":"%(filename)s", 
-#     "logRecordCreationTime":"%(created)f", 
-#     "functionName":"%(funcName)s", 
-#     "levelNo":"%(levelno)s", 
-#     "lineNo":"%(lineno)d", 
-#     "time":"%(msecs)d", 
-#     "levelName":"%(levelname)s", 
-#     "message":"%(message)s"}',
-# },
-
 class LogstashFormatter(LogstashFormatterVersion1):
     def __init__(self, **kwargs):
         ret = super(LogstashFormatter, self).__init__(**kwargs)
         self.host_id = settings.CLUSTER_HOST_ID
         return ret
 
+    def reformat_data_for_log(self, raw_data, kind=None):
+        '''
+        Process dictionaries from various contexts (job events, activity stream
+        changes, etc.) to give meaningful information
+        Output a dictionary which will be passed in logstash or syslog format
+        to the logging receiver
+        '''
+        if kind == 'activity_stream':
+            return raw_data
+        rename_fields = set((
+            'args', 'asctime', 'created', 'exc_info', 'exc_text', 'filename',
+            'funcName', 'id', 'levelname', 'levelno', 'lineno', 'module',
+            'msecs', 'msecs', 'message', 'msg', 'name', 'pathname', 'process',
+            'processName', 'relativeCreated', 'thread', 'threadName', 'extra',
+            'auth_token', 'tags', 'host', 'host_id', 'level', 'port', 'uuid'))
+        if kind == 'system_tracking':
+            data = copy(raw_data['facts_data'])
+        elif kind == 'job_events':
+            data = copy(raw_data['event_model_data'])
+        else:
+            data = copy(raw_data)
+        if isinstance(data, basestring):
+            data = json.loads(data)
+        skip_fields = ('res', 'password', 'event_data', 'stdout')
+        data_for_log = {}
+
+        def index_by_name(alist):
+            """Takes a list of dictionaries with `name` as a key in each dict
+            and returns a dictionary indexed by those names"""
+            adict = {}
+            for item in alist:
+                subdict = copy(item)
+                if 'name' in subdict:
+                    name = subdict.get('name', None)
+                elif 'path' in subdict:
+                    name = subdict.get('path', None)
+                if name:
+                    # Logstash v2 can not accept '.' in a name
+                    name = name.replace('.', '_')
+                    adict[name] = subdict
+            return adict
+
+        if kind == 'job_events':
+            data.update(data.get('event_data', {}))
+            for fd in data:
+                if fd in skip_fields:
+                    continue
+                key = fd
+                if fd in rename_fields:
+                    key = 'event_%s' % fd
+                val = data[fd]
+                if key.endswith('created'):
+                    time_float = time.mktime(data[fd].timetuple())
+                    val = self.format_timestamp(time_float)
+                data_for_log[key] = val
+        elif kind == 'system_tracking':
+            module_name = raw_data['module_name']
+            if module_name in ['services', 'packages', 'files']:
+                data_for_log[module_name] = index_by_name(data)
+            elif module_name == 'ansible':
+                data_for_log['ansible'] = data
+                # Remove sub-keys with data type conflicts in elastic search
+                data_for_log['ansible'].pop('ansible_python_version', None)
+                data_for_log['ansible']['ansible_python'].pop('version_info', None)
+            else:
+                data_for_log['facts'] = data
+            data_for_log['module_name'] = module_name
+        return data_for_log
+
     def get_extra_fields(self, record):
         fields = super(LogstashFormatter, self).get_extra_fields(record)
-        fields['cluster_host_id'] = self.host_id
+        if record.name.startswith('awx.analytics'):
+            log_kind = record.name.split('.')[-1]
+            fields = self.reformat_data_for_log(fields, kind=log_kind)
         return fields
 
     def format(self, record):
-        # Create message dict
-        # message = record.getMessage()
-        # print ' message ' + str(message)
         message = {
+            # Fields not included, but exist in related logs
+            # 'path': record.pathname
+            # '@version': '1', # from python-logstash
+            # 'tags': self.tags,
             '@timestamp': self.format_timestamp(record.created),
-            '@version': '1',
             'message': record.getMessage(),
             'host': self.host,
-            'path': record.pathname,
-            'tags': self.tags,
             'type': self.message_type,
 
             # Extra Fields
             'level': record.levelname,
             'logger_name': record.name,
+            'cluster_host_id': self.host_id
         }
 
         # Add extra fields

awx/main/log_utils/handlers.py

@@ -24,12 +24,11 @@ from requests_futures.sessions import FuturesSession
 from logstash import formatter
 
 # custom
+from copy import copy
 from requests.auth import HTTPBasicAuth
 from django.conf import settings as django_settings
 
 
-ENABLED_LOGS = ['ansible']
-
 # Logstash
 # https://github.com/vklochan/python-logstash
 class TCPLogstashHandler(logging.handlers.SocketHandler, object):
@@ -64,14 +63,16 @@ PARAM_NAMES = {
     'message_type': 'LOG_AGGREGATOR_TYPE',
     'username': 'LOG_AGGREGATOR_USERNAME',
     'password': 'LOG_AGGREGATOR_PASSWORD',
+    'enabled_loggers': 'LOG_AGGREGATOR_LOGGERS',
+    'indv_facts': 'LOG_AGGREGATOR_INDIVIDUAL_FACTS',
 }
 # TODO: figure out what to do with LOG_AGGREGATOR_LOGGERS (if anything)
 
 
-def bg_cb(sess, resp):
+def unused_callback(sess, resp):
-    """ Don't do anything with the response """
     pass
 
+
 class HTTPSHandler(logging.Handler):
     def __init__(self, fqdn=False, **kwargs):
         super(HTTPSHandler, self).__init__()
@@ -83,7 +84,6 @@ class HTTPSHandler(logging.Handler):
             if settings_val:
                 setattr(self, fd, settings_val)
             elif fd in kwargs:
-                attr_name = fd
                 setattr(self, fd, kwargs[fd])
             else:
                 setattr(self, fd, None)
@@ -100,37 +100,53 @@ class HTTPSHandler(logging.Handler):
         if self.message_type == 'logstash':
             if not self.username:
                 # Logstash authentication not enabled
-                return kwargs
+                return
             logstash_auth = HTTPBasicAuth(self.username, self.password)
             self.session.auth = logstash_auth
         elif self.message_type == 'splunk':
-            ## Auth used by Splunk logger library
+            auth_header = "Splunk %s" % self.password
-            # self.session.auth = ('x', self.access_token)
+            headers = {
-            # self.session.headers.update({'Content-Encoding': 'gzip'})
+                "Authorization": auth_header,
-            auth_header = "Splunk %s" % self.token
+                "Content-Type": "application/json"
-            headers = dict(Authorization=auth_header)
+            }
             self.session.headers.update(headers)
 
     def emit(self, record):
+        if (self.host == '' or self.enabled_loggers is None or
+                record.name.split('.')[-1] not in self.enabled_loggers):
+            return
         try:
             payload = self.format(record)
-            # TODO: move this enablement logic to rely on individual loggers once
-            # the enablement config variable is hooked up
-            payload_data = json.loads(payload)
-            if payload_data['logger_name'].startswith('awx.analytics.system_tracking'):
-                st_type = None
-                for fd in ['services', 'packages', 'files', 'ansible']:
-                    if fd in payload_data:
-                        st_type = fd
-                        break
-                if st_type not in ENABLED_LOGS:
-                    return
             host = self.host
             if not host.startswith('http'):
                 host = 'http://%s' % self.host
-            if self.port != 80:
+            if self.port != 80 and self.port is not None:
                 host = '%s:%s' % (host, str(self.port))
-            kwargs = dict(data=payload, background_callback=bg_cb)
+
+            # Special action for System Tracking, queue up multiple log messages
+            if self.indv_facts:
+                payload_data = json.loads(payload)
+                if record.name.startswith('awx.analytics.system_tracking'):
+                    module_name = payload_data['module_name']
+                    if module_name in ['services', 'packages', 'files']:
+                        facts_dict = payload_data.pop(module_name)
+                        for key in facts_dict:
+                            fact_payload = copy(payload_data)
+                            fact_payload.update(facts_dict[key])
+                            kwargs = dict(data=json.dumps(fact_payload), background_callback=unused_callback)
+                            self.session.post(host, **kwargs)
+                        return
+
+            kwargs = dict(data=payload, background_callback=unused_callback)
+            # # splunk doesn't take "@" in the keys
+            # if self.message_type == 'splunk':
+            #     payload_dict = json.loads(payload)
+            #     new_dict = {}
+            #     for key in payload_dict:
+            #         new_key = key.replace('@', '')
+            #         new_dict[new_key] = payload_dict[key]
+            #     new_payload = json.dumps(new_dict)
+            #     kwargs['data'] = json.dumps(new_dict)
             self.session.post(host, **kwargs)
         except (KeyboardInterrupt, SystemExit):
             raise
@@ -162,9 +178,7 @@ class SplunkLogger(logging.Handler):
     def _set_auth(self, access_token, project_id, api_domain):
         # The access token and project id passed as parameter override the ones
         # configured in the .splunk_logger file.
-        if access_token is not None\
+        if access_token is not None and project_id is not None and api_domain is not None:
-        and project_id is not None\
-        and api_domain is not None:
             self.project_id = project_id
             self.access_token = access_token
             self.api_domain = api_domain
@@ -173,9 +187,7 @@ class SplunkLogger(logging.Handler):
             # Try to get the credentials form the configuration file
             self.project_id, self.access_token, self.api_domain = parse_config_file()
 
-            if self.project_id is None\
+            if self.project_id is None or self.access_token is None or self.api_domain is None:
-            or self.access_token is None\
-            or self.api_domain is None:
                 # Try to get the credentials form the environment variables
                 self.project_id, self.access_token, self.api_domain = get_config_from_env()
 

awx/main/management/commands/run_fact_cache_receiver.py

@@ -16,10 +16,9 @@ from django.utils import timezone
 # AWX
 from awx.main.models.fact import Fact
 from awx.main.models.inventory import Host
-from awx.main.utils import format_for_log
 
 logger = logging.getLogger('awx.main.commands.run_fact_cache_receiver')
-data_logger = logging.getLogger('awx.analytics.system_tracking')
+analytics_logger = logging.getLogger('awx.analytics.system_tracking')
 
 
 class FactBrokerWorker(ConsumerMixin):
@@ -83,7 +82,8 @@ class FactBrokerWorker(ConsumerMixin):
             # Create new Fact entry
             fact_obj = Fact.add_fact(host_obj.id, module_name, self.timestamp, facts)
             logger.info('Created new fact <fact_id, module> <%s, %s>' % (fact_obj.id, module_name))
-            data_logger.info('Received message with fact data', extra=format_for_log({module_name: facts}, kind="fact"))
+            analytics_logger.info('Received message with fact data', extra=dict(
+                module_name=module_name, facts_data=facts))
         return fact_obj
 
 

awx/main/models/jobs.py

@@ -32,7 +32,6 @@ from awx.main.models.notifications import (
 from awx.main.utils import (
     ignore_inventory_computed_fields,
     parse_yaml_or_json,
-    format_for_log
 )
 from awx.main.redact import PlainTextCleaner
 from awx.main.fields import ImplicitRoleField
@@ -44,7 +43,7 @@ from awx.main.consumers import emit_channel_notification
 
 
 logger = logging.getLogger('awx.main.models.jobs')
-event_logger = logging.getLogger('awx.analytics.job_events')
+analytics_logger = logging.getLogger('awx.analytics.job_events')
 
 __all__ = ['JobTemplate', 'Job', 'JobHostSummary', 'JobEvent', 'SystemJobOptions', 'SystemJobTemplate', 'SystemJob']
 
@@ -1188,8 +1187,7 @@ class JobEvent(CreatedModifiedModel):
             if parent_id:
                 kwargs['parent_id'] = parent_id
 
-        # event_logger.info('Body: {}'.format(str(data_for_log)), extra=data_for_log)
+        analytics_logger.info('Job event data saved.', extra=dict(event_model_data=kwargs))
-        event_logger.info('Job event data saved.', extra=format_for_log(kwargs, kind='event'))
 
         job_event = JobEvent.objects.create(**kwargs)
 
@@ -1201,6 +1199,7 @@ class JobEvent(CreatedModifiedModel):
         # Save artifact data to parent job (if provided).
         if artifact_data:
             artifact_dict = json.loads(artifact_data)
+            event_data = kwargs.get('event_data', None)
             if event_data and isinstance(event_data, dict):
                 res = event_data.get('res', None)
                 if res and isinstance(res, dict):

awx/main/signals.py

@@ -20,7 +20,7 @@ from crum.signals import current_user_getter
 from awx.main.models import * # noqa
 from awx.api.serializers import * # noqa
 from awx.main.utils import model_instance_diff, model_to_dict, camelcase_to_underscore
-from awx.main.utils import ignore_inventory_computed_fields, ignore_inventory_group_removal, _inventory_updates, format_for_log
+from awx.main.utils import ignore_inventory_computed_fields, ignore_inventory_group_removal, _inventory_updates
 from awx.main.tasks import update_inventory_computed_fields
 
 from awx.main.consumers import emit_channel_notification
@@ -28,7 +28,6 @@ from awx.main.consumers import emit_channel_notification
 __all__ = []
 
 logger = logging.getLogger('awx.main.signals')
-analytics_logger = logging.getLogger('awx.analytics.activity_stream')
 
 # Update has_active_failures for inventory/groups when a Host/Group is deleted,
 # when a Host-Group or Group-Group relationship is updated, or when a Job is deleted
@@ -370,15 +369,11 @@ def activity_stream_create(sender, instance, created, **kwargs):
         if type(instance) == Job:
             if 'extra_vars' in changes:
                 changes['extra_vars'] = instance.display_extra_vars()
-        changes_dict = json.dumps(changes)
         activity_entry = ActivityStream(
             operation='create',
             object1=object1,
-            changes=changes_dict)
+            changes=json.dumps(changes))
         activity_entry.save()
-        # analytics_logger.info('Activity Stream create entry for %s' % str(object1),
-        #                       extra=format_for_log(changes, kind='activity_stream',
-        #                       actor=activity_entry.actor, operation='update', object1=object1))
         #TODO: Weird situation where cascade SETNULL doesn't work
         #      it might actually be a good idea to remove all of these FK references since
         #      we don't really use them anyway.
@@ -406,9 +401,6 @@ def activity_stream_update(sender, instance, **kwargs):
         object1=object1,
         changes=json.dumps(changes))
     activity_entry.save()
-    # analytics_logger.info('Activity Stream update entry for %s' % str(object1),
-    #                       extra=format_for_log(changes, kind='activity_stream',
-    #                       actor=activity_entry.actor, operation='update', object1=object1))
     if instance._meta.model_name != 'setting':  # Is not conf.Setting instance
         getattr(activity_entry, object1).add(instance)
 

awx/main/tests/unit/test_log_formatter.py

@@ -6,7 +6,8 @@ event_data = {
     'stdout': u'\x1b[0;36mskipping: [host1]\x1b[0m\r\n', u'uuid': u'ffe4858c-ac38-4cab-9192-b07bdbe80502',
     u'created': datetime.datetime(2016, 11, 10, 14, 59, 16, 376051), 'counter': 17, u'job_id': 209,
     u'event': u'runner_on_skipped', 'parent_id': 1937, 'end_line': 24, 'start_line': 23,
-    u'event_data': {u'play_pattern': u'all', u'play': u'all', u'task': u'Scan files (Windows)',
+    u'event_data': {
+        u'play_pattern': u'all', u'play': u'all', u'task': u'Scan files (Windows)',
         u'task_args': u'paths={{ scan_file_paths }}, recursive={{ scan_use_recursive }}, get_checksum={{ scan_use_checksum }}',
         u'remote_addr': u'host1', u'pid': 1427, u'play_uuid': u'da784361-3811-4ea7-9cc8-46ec758fde66',
         u'task_uuid': u'4f9525fd-bc25-4ace-9eb2-adad9fa21a94', u'event_loop': None,

awx/main/utils.py

@@ -17,7 +17,6 @@ import urlparse
 import threading
 import contextlib
 import tempfile
-from copy import copy
 
 # Decorator
 from decorator import decorator
@@ -825,64 +824,3 @@ class OutputEventFilter(object):
             self._current_event_data = next_event_data
         else:
             self._current_event_data = None
-
-def format_for_log(raw_data, kind=None, **kwargs):
-    '''
-    Process dictionaries from various contexts (job events, activity stream
-    changes, etc.) to give meaningful information
-    Output a dictionary which will be passed in logstash or syslog format
-    to the logging receiver
-    '''
-    rename_fields = set((
-        'args', 'asctime', 'created', 'exc_info', 'exc_text', 'filename',
-        'funcName', 'id', 'levelname', 'levelno', 'lineno', 'module',
-        'msecs', 'msecs', 'message', 'msg', 'name', 'pathname', 'process',
-        'processName', 'relativeCreated', 'thread', 'threadName', 'extra',
-        'auth_token', 'tags', 'host', 'host_id', 'level', 'port', 'uuid'))
-    data = copy(raw_data)
-    if isinstance(data, basestring):
-        data = json.loads(data)
-    data.update(kwargs)
-    skip_fields = ('res', 'password', 'event_data', 'stdout')
-    data_for_log = {}
-
-    def index_by_name(alist):
-        """Takes a list of dictionaries with `name` as a key in each dict
-        and returns a dictionary indexed by those names"""
-        adict = {}
-        for item in alist:
-            subdict = copy(item)
-            name = subdict.pop('name', None)
-            if name:
-                # Logstash v2 can not accept '.' in a name
-                name = name.replace('.', '_')
-                adict[name] = subdict
-        return adict
-
-    if kind == 'event':
-        data.update(data.get('event_data', {}))
-        for fd in data:
-            if fd in skip_fields:
-                continue
-            key = fd
-            if fd in rename_fields:
-                key = 'event_%s' % fd
-            if type(data[fd]) is dict:
-                data_for_log[key] = len(data[fd])
-            else:
-                data_for_log[key] = data[fd]
-    elif kind == 'fact':
-        if 'services' in data:
-            data_for_log['services'] = index_by_name(data['services'])
-        elif 'packages' in data:
-            data_for_log['packages'] = index_by_name(data['packages'])
-        elif 'files' in data:
-            data_for_log['files'] = index_by_name(data['files'])
-        elif 'ansible' in data:
-            data_for_log['ansible'] = data['ansible']
-            # Remove sub-keys with data type conflicts in elastic search
-            data_for_log['ansible'].pop('ansible_python_version', None)
-            data_for_log['ansible']['ansible_python'].pop('version_info', None)
-        else:
-            data_for_log['facts'] = data
-    return data_for_log

awx/settings/defaults.py

@@ -842,19 +842,9 @@ LOGGING = {
         'simple': {
             'format': '%(asctime)s %(levelname)-8s %(name)s %(message)s',
         },
-        'logstash': {
-            '()': 'awx.main.log_utils.formatters.LogstashFormatter'
-        },
         'json': {
             '()': 'awx.main.log_utils.formatters.LogstashFormatter'
         }
-        # From loggly examples
-        # 'json': {
-        #     'format': '{ "loggerName":"%(name)s", "asciTime":"%(asctime)s", "fileName":"%(filename)s", "logRecordCreationTime":"%(created)f", "functionName":"%(funcName)s", "levelNo":"%(levelno)s", "lineNo":"%(lineno)d", "time":"%(msecs)d", "levelName":"%(levelname)s", "message":"%(message)s"}',
-        # },
-        # 'json': { 
-        #     'format': '{"message": %(message)s}',
-        # },
     },
     'handlers': {
         'console': {
@@ -882,20 +872,6 @@ LOGGING = {
             'formatter': 'json',
             'host': '',
         },
-        'logstash': {
-            'level': 'INFO',
-            'class': 'awx.main.log_utils.handlers.HTTPSHandler',
-            'host': 'logstash',   # IP/name of our Logstash EC2 instance
-            'port': 8085,
-            # 'port': 5000,
-            # 'version': 1,
-            'message_type': 'logstash',
-            'fqdn': True,
-            # 'tags': ['tower'],
-            'formatter': 'json',
-            'username': 'awx_logger',
-            'password': 'workflows',
-        },
         'mail_admins': {
             'level': 'ERROR',
             'filters': ['require_debug_false'],