def get_latest_stat(node, resource):
try:
stats = NS.time_series_db_manager.get_plugin().get_metric_stats(
node, resource, 'latest')
if stats == "[]" or not stats:
raise TendrlPerformanceMonitoringException(
'Stats not yet available in time series db')
stat = re.search('Current:(.+?)Max', stats)
if not stat:
raise TendrlPerformanceMonitoringException(
'Failed to get latest stat of %s of node %s for summary'
'Error: Current utilization not found' % (resource, node))
stat = re.search('Current:(.+?)Max', stats).group(1)
if math.isnan(float(stat)):
raise TendrlPerformanceMonitoringException(
'Received nan for utilization %s of %s' % (resource, node))
return float(stat)
except (ValueError, urllib3.exceptions.HTTPError,
TendrlPerformanceMonitoringException) as ex:
Event(
ExceptionMessage(priority="debug",
publisher=NS.publisher_id,
payload={
"message":
'Failed to get latest stat of %s of '
'node %s for node summary.' %
(resource, node),
"exception":
ex
}))
raise ex
def get_aggregated_stats(self,
aggregation_type,
entity_names,
metric_name,
time_interval=None,
start_time=None,
end_time=None):
target = ''
for entity_name in entity_names:
target = '%s%s.%s.%s,' % (target, self.prefix,
entity_name.replace('.',
'_'), metric_name)
target = target[:-1]
if aggregation_type == pm_consts.AVERAGE:
target = 'averageSeries(%s)' % target
if time_interval:
if time_interval == 'latest':
target = "cactiStyle(%s)" % target
else:
start_time = self.parse_time(time_interval)
if start_time:
start_time = self.parse_time(start_time)
if end_time:
end_time = self.parse_time(end_time)
url = 'http://%s:%s/render?target=%s&format=json' % (
self.host, str(self.port), target)
if start_time:
url = "%s&from=%s" % (url, start_time)
if end_time:
url = "%s&until=%s" % (url, end_time)
try:
stats = self.http.request('GET', url, timeout=5)
if stats.status == 200:
# TODO(Anmol): remove nulls from graphite data before returning
# data. Explore the possibility of achieving this using some
# tuning factor in graphite.
data = re.sub('\[null, [0-9]+\], ', '', stats.data)
data = re.sub(', \[null, [0-9]+\]', '', data)
return data
else:
TendrlPerformanceMonitoringException(
'Request status code: %s' % str(stats.status))
except (ValueError, Exception) as ex:
Event(
ExceptionMessage(priority="debug",
publisher=NS.publisher_id,
payload={
"message":
'Failed to fetch stats for metric %s'
' of %s using url. %s' %
(metric_name, entity_name, url),
"exception":
ex
}))
raise TendrlPerformanceMonitoringException(str(ex))
def get_node_disk_iops_stats(self,
node_id,
time_interval=None,
start_time=None,
end_time=None):
node_name = central_store_util.get_node_name_from_id(node_id)
node_name = node_name.replace('.', '_')
target = Template(
'sumSeries(averageSeries($prefix.$node_name.disk-*.disk_ops.write'
'), averageSeries($prefix.$node_name.disk-*.disk_ops.read))'
).substitute(
prefix=self.prefix,
node_name=node_name,
)
target = urllib.quote(target)
if time_interval:
if time_interval == 'latest':
target = "cactiStyle(%s)" % target
else:
start_time = self.parse_time(time_interval)
if start_time:
start_time = self.parse_time(start_time)
if end_time:
end_time = self.parse_time(end_time)
url = 'http://%s:%s/render?target=%s&format=json' % (
self.host, str(self.port), target)
if start_time:
url = "%s&from=%s" % (url, start_time)
if end_time:
url = "%s&until=%s" % (url, end_time)
try:
stats = self.http.request('GET', url, timeout=5)
if stats.status == 200:
# TODO(Anmol): remove nulls from graphite data before returning
# data. Explore the possibility of achieving this using some
# tuning factor in graphite.
data = re.sub('\[null, [0-9]+\], ', '', stats.data)
data = re.sub(', \[null, [0-9]+\]', '', data)
return data
else:
TendrlPerformanceMonitoringException(
'Request status code: %s' % str(stats.status))
except (ValueError, Exception) as ex:
Event(
ExceptionMessage(priority="debug",
publisher=NS.publisher_id,
payload={
"message":
'Failed to fetch %s stats using url %s'
'. Error %s' % (target, url),
"exception":
ex
}))
raise TendrlPerformanceMonitoringException(str(ex))
def get_cluster_summary(cluster_id):
try:
summary = ClusterSummary(cluster_id=cluster_id)
if not summary.exists():
raise TendrlPerformanceMonitoringException(
"No summary found for cluster %s" % cluster_id)
summary = summary.load().to_json()
for key, value in summary.items():
if (key.startswith("_")
or key in ['hash', 'updated_at', 'value', 'list']):
del summary[key]
return summary
except Exception as ex:
raise TendrlPerformanceMonitoringException(str(ex))
def get_system_summary(cluster_type):
try:
summary = SystemSummary(sds_type=cluster_type)
if not summary.exists():
raise TendrlPerformanceMonitoringException(
"No clusters of type %s found" % cluster_type)
summary = summary.load().to_json()
for key, value in summary.items():
if (key.startswith("_")
or key in ['hash', 'updated_at', 'value', 'list']):
del summary[key]
return summary
except Exception as ex:
raise TendrlPerformanceMonitoringException(str(ex))
def get_most_used_bricks(self, bricks):
brick_utilizations = []
if not bricks:
return brick_utilizations
for brick_path, brick_det in bricks.iteritems():
if (
'utilization' in brick_det and
'used_percent' in brick_det['utilization']
):
brick_utilizations.append(brick_det['utilization'])
else:
Event(
ExceptionMessage(
priority="debug",
publisher=NS.publisher_id,
payload={
"message": "No utilization info for brick "
"%s" % brick_path,
"exception": TendrlPerformanceMonitoringException(
'No utilization info for brick %s' % brick_path
)
}
)
)
brick_utilizations = sorted(
brick_utilizations,
key=lambda k: k['used_percent']
)
brick_utilizations.reverse()
return brick_utilizations[:5]
def get_metrics(self, entity_name):
url = 'http://%s:%s/metrics/index.json' % (self.host, str(self.port))
try:
gevent.sleep(5)
resp = self.http.request('GET', url, timeout=5)
if resp.status != 200:
raise TendrlPerformanceMonitoringException(
'Request status code: %s' % str(resp.status_code))
data = resp.data
metrics = ast.literal_eval(data)
result = []
prefix = "%s.%s." % (self.prefix, entity_name.replace('.', '_'))
split_metrics = []
for metric in metrics:
if metric.startswith(prefix):
split_metrics = metric.split(prefix)
result.append(split_metrics[1])
return str(result)
except (ValueError, Exception) as ex:
Event(
ExceptionMessage(
priority="debug",
publisher=NS.publisher_id,
payload={
"message":
'Failed to get metrics for %s.' % entity_name,
"exception": ex
}))
def get_node_name_from_id(node_id):
try:
node_name_path = '/nodes/%s/NodeContext/fqdn' % node_id
return NS._int.client.read(node_name_path).value
except (EtcdKeyNotFound, EtcdConnectionFailed, ValueError, SyntaxError,
EtcdException, TypeError) as ex:
raise TendrlPerformanceMonitoringException(str(ex))
def initiate_config_generation(node_det):
try:
job_params = {
'node_ids': [node_det.get('node_id')],
"run": 'node_monitoring.flows.ConfigureCollectd',
'type': 'monitoring',
"parameters": {
'plugin_name':
node_det['plugin'],
'plugin_conf_params':
json.dumps(node_det['plugin_conf']).encode('utf-8'),
'Node.fqdn':
node_det['fqdn'],
'Service.name':
'collectd',
},
}
Job(
job_id=str(uuid.uuid4()),
status='new',
payload=job_params,
).save()
except (EtcdException, EtcdConnectionFailed, Exception) as ex:
raise TendrlPerformanceMonitoringException(
'Failed to intiate monitoring configuration for plugin \
%s on %s with parameters %s.Error %s' %
(node_det['plugin'], node_det['fqdn'],
json.dumps(node_det['plugin_conf']), str(ex)))
def get_latest_stats(node, resource):
try:
node_name = central_store_util.get_node_name_from_id(node)
stats = NS.time_series_db_manager.get_plugin().get_metric_stats(
node_name,
resource,
'latest'
)
if stats == "[]" or not stats:
raise TendrlPerformanceMonitoringException(
'Stats not yet available in time series db'
)
return re.findall('Current:(.+?)Max', stats)
except TendrlPerformanceMonitoringException as ex:
Event(
ExceptionMessage(
priority="debug",
publisher=NS.publisher_id,
payload={"message": 'Failed to get latest stats of %s of '
'node %s for node summary.'
% (resource, node),
"exception": ex
}
)
)
raise ex
def get_node_summary():
try:
# Only 1 filter that is the node list is the only supported filter
# anything else is simply ignored.
summary = []
ret_code = 200
exs = ''
is_filter = (len(request.args) == 1
and request.args.items()[0][0] == 'node_ids')
if is_filter:
node_list = (request.args.items()[0][1]).split(",")
for index, node in enumerate(node_list):
uuid_string = node_list[index].strip()
if UUID(uuid_string,
version=4).hex == uuid_string.replace('-', ''):
node_list[index] = node_list[index].strip()
else:
raise TendrlPerformanceMonitoringException(
'Node id %s in the parameter is not a valid uuid' %
(uuid_string))
summary, ret_code, exs = \
central_store_util.get_node_summary(node_list)
else:
summary, ret_code, exs = \
central_store_util.get_node_summary()
return Response(json.dumps(summary),
status=ret_code,
mimetype='application/json')
except (etcd.EtcdKeyNotFound, etcd.EtcdConnectionFailed, ValueError,
SyntaxError, etcd.EtcdException,
TendrlPerformanceMonitoringException, TypeError) as ex:
return Response(str(ex), status=500, mimetype='application/json')
def get_node_role(node_id):
try:
return NS._int.client.read('/nodes/%s/NodeContext/tags' %
node_id).value
except Exception as ex:
raise TendrlPerformanceMonitoringException(
"Failed to fetch the role of node %s. Error %s" %
(node_id, str(ex)))
def get_node_cluster_name(node_id):
try:
return NS._int.client.read('/nodes/%s/TendrlContext/cluster_name' %
node_id).value
except Exception as ex:
raise TendrlPerformanceMonitoringException(
"Failed to fetch cluster name for node %s. Error: %s" %
(node_id, str(ex)))
def get_node_ids(self):
try:
node_ids = []
nodes_etcd = tendrl_ns.etcd_orm.client.read('/nodes')
for node in nodes_etcd._children:
node_ids.append(node['key'][len('/nodes/'):])
return node_ids
except EtcdKeyNotFound:
return []
except (EtcdConnectionFailed, ValueError, SyntaxError,
TypeError) as ex:
raise TendrlPerformanceMonitoringException(str(ex))
def get_metric_stats(self, entity_name, metric_name, time_interval=None):
metric_name = '%s.%s' % (entity_name.replace('.', '_'), metric_name)
target = '%s.%s' % (self.prefix, metric_name)
if time_interval == 'latest':
target = "cactiStyle(%s)" % target
url = 'http://%s:%s/render?target=%s&format=json' % (
self.host, str(self.port), target)
try:
stats = self.http.request('GET', url, timeout=5)
if stats.status == 200:
return stats.data
else:
TendrlPerformanceMonitoringException(
'Request status code: %s' % str(
data.status_code
)
)
except (ValueError, Exception) as ex:
LOG.error('Failed to fetch stats for metric %s of %s using url %s.Error %s ' % (
metric_name, entity_name, url, str(ex)), exc_info=True)
raise TendrlPerformanceMonitoringException(str(ex))
def get_cluster_ids():
try:
cluster_ids = []
clusters_etcd = NS._int.client.read('/clusters')
for cluster in clusters_etcd.leaves:
cluster_key_contents = cluster.key.split('/')
if len(cluster_key_contents) == 3:
cluster_ids.append(cluster_key_contents[2])
return cluster_ids
except EtcdKeyNotFound:
return []
except (EtcdConnectionFailed, ValueError, SyntaxError, TypeError) as ex:
raise TendrlPerformanceMonitoringException(str(ex))
def get_node_alert_ids(node_id=None):
alert_ids = []
try:
alerts = NS._int.client.read('/alerting/nodes/%s' % node_id)
for alert in alerts.leaves:
key_contents = alert.key.split('/')
if len(key_contents) == 5:
alert_ids.append(key_contents[4])
except EtcdKeyNotFound as ex:
return alert_ids
except (EtcdConnectionFailed, EtcdException) as ex:
raise TendrlPerformanceMonitoringException(str(ex))
return alert_ids
def get_node_ids():
try:
node_ids = []
nodes_etcd = NS._int.client.read('/nodes')
for node in nodes_etcd.leaves:
node_key_contents = node.key.split('/')
if len(node_key_contents) == 3:
node_ids.append(node_key_contents[2])
return node_ids
except EtcdKeyNotFound:
return []
except (EtcdConnectionFailed, ValueError, SyntaxError, TypeError) as ex:
raise TendrlPerformanceMonitoringException(str(ex))
def get_sdsthroughput(sds_type, network_type):
try:
start_time = None
end_time = None
time_interval = None
if len(request.args.items()) > 0:
for request_param in request.args.items():
if request_param[0] == "start_time":
start_time = request_param[1]
elif request_param[0] == "end_time":
end_time = request_param[1]
elif request_param[0] == "interval":
time_interval = request_param[1]
# validate sds-type
if sds_type not in NS.sds_monitoring_manager.supported_sds:
raise TendrlPerformanceMonitoringException(
'Unsupported sds %s' % sds_type
)
entity_name, metric_name = NS.time_series_db_manager.\
get_timeseriesnamefromresource(
sds_type=sds_type,
network_type=network_type,
resource_name=pm_consts.SYSTEM_THROUGHPUT,
utilization_type=pm_consts.USED
).split(
NS.time_series_db_manager.get_plugin().get_delimeter(),
1
)
return Response(
NS.time_series_db_manager.\
get_plugin().\
get_metric_stats(
entity_name,
metric_name,
time_interval=time_interval,
start_time=start_time,
end_time=end_time
),
status=200,
mimetype='application/json'
)
except (
ValueError,
etcd.EtcdKeyNotFound,
etcd.EtcdConnectionFailed,
SyntaxError,
etcd.EtcdException,
TypeError,
TendrlPerformanceMonitoringException
) as ex:
return Response(str(ex), status=500, mimetype='application/json')
def get_sdsutilization(sds_type, utiliation_type):
try:
start_time = None
end_time = None
time_interval = None
if len(request.args.items()) > 0:
for request_param in request.args.items():
if request_param[0] == "start_time":
start_time = request_param[1]
elif request_param[0] == "end_time":
end_time = request_param[1]
elif request_param[0] == "interval":
time_interval = request_param[1]
# validate sds-type
if sds_type not in NS.sds_monitoring_manager.supported_sds:
raise TendrlPerformanceMonitoringException(
'Unsupported sds %s' % sds_type
)
entity_name, metric_name = NS.time_series_db_manager.\
get_timeseriesnamefromresource(
resource_name=pm_consts.SYSTEM_UTILIZATION,
utilization_type=utiliation_type,
sds_type=sds_type
).split(
NS.time_series_db_manager.get_plugin().get_delimeter(),
1
)
return Response(
NS.time_series_db_manager.\
get_plugin().\
get_metric_stats(
entity_name,
metric_name,
time_interval=time_interval,
start_time=start_time,
end_time=end_time
),
status=200,
mimetype='application/json'
)
except (
AttributeError,
ValueError,
etcd.EtcdException,
SyntaxError,
urllib3.exceptions.HTTPError,
TypeError,
TendrlPerformanceMonitoringException
) as ex:
return Response(str(ex), status=500, mimetype='application/json')
def get_system_summary(cluster_type):
try:
if cluster_type not in NS.sds_monitoring_manager.supported_sds:
raise TendrlPerformanceMonitoringException('Unsupported sds %s' %
cluster_type)
summary = central_store_util.get_system_summary(cluster_type)
return Response(json.dumps(summary),
status=200,
mimetype='application/json')
except TendrlPerformanceMonitoringException as ex:
return Response('Failed to fetch %s system summary.Error %s' %
(cluster_type, str(ex)),
status=500,
mimetype='application/json')
def get_clusters_iops():
try:
cluster_list = None
start_time = None
end_time = None
time_interval = None
if len(request.args.items()) > 0:
for request_param in request.args.items():
if request_param[0] == "start_time":
start_time = request_param[1]
elif request_param[0] == "end_time":
end_time = request_param[1]
elif request_param[0] == "interval":
time_interval = request_param[1]
elif request_param[0] == "cluster_ids":
cluster_list = (request.args.items()[0][1]).split(",")
iops = []
ret_code = 200
exs = ''
if cluster_list:
for index, node in enumerate(cluster_list):
uuid_string = cluster_list[index].strip()
if UUID(uuid_string,
version=4).hex == uuid_string.replace('-', ''):
cluster_list[index] = cluster_list[index].strip()
else:
raise TendrlPerformanceMonitoringException(
'Cluster id %s in the parameter is not a valid '
'uuid' % (uuid_string))
iops, ret_code, exs = \
central_store_util.get_cluster_iops(
cluster_list,
time_interval=time_interval,
start_time=start_time,
end_time=end_time
)
else:
iops, ret_code, exs = \
central_store_util.get_cluster_iops(
time_interval=time_interval,
start_time=start_time,
end_time=end_time
)
return Response(json.dumps(iops),
status=ret_code,
mimetype='application/json')
except (etcd.EtcdKeyNotFound, etcd.EtcdConnectionFailed, ValueError,
SyntaxError, etcd.EtcdException,
TendrlPerformanceMonitoringException, TypeError) as ex:
return Response(str(ex), status=500, mimetype='application/json')
def get_configs(self):
# TODO(Anmol) : Attempt reading:
# /_tendrl/config/performance_monitoring/clusters/{cluster-id} and if
# not already present, default back to defaults in:
# /_tendrl/config/performance_monitoring
try:
configs = ''
conf = tendrl_ns.etcd_orm.client.read(
'/_tendrl/config/performance_monitoring')
configs = conf.value
return yaml.safe_load(configs)
except (EtcdKeyNotFound, EtcdConnectionFailed, ValueError, SyntaxError,
EtcdException) as ex:
LOG.error('Fetching monitoring configurations failed. Error %s' %
ex)
raise TendrlPerformanceMonitoringException(str(ex))
def get_nodes_details():
nodes_dets = []
try:
nodes = NS._int.client.read('/nodes/')
for node in nodes.leaves:
if node.key.startswith('/nodes/'):
node_id = (node.key.split('/')[2]).encode('ascii', 'ignore')
fqdn = (NS._int.client.read('/nodes/%s/NodeContext/fqdn' %
(node_id)).value).encode(
'ascii', 'ignore')
nodes_dets.append({'node_id': node_id, 'fqdn': fqdn})
return nodes_dets
except EtcdKeyNotFound:
return nodes_dets
except EtcdConnectionFailed as ex:
raise TendrlPerformanceMonitoringException(str(ex))
def get_nodes_details(self):
nodes_dets = []
try:
nodes = tendrl_ns.etcd_orm.client.read('/nodes/', recursive=True)
for node in nodes._children:
if node['key'].startswith('/nodes/'):
node_id = (node['key'][len('/nodes/'):]).encode(
'ascii', 'ignore')
fqdn = (tendrl_ns.etcd_orm.client.read(
'%s/NodeContext/fqdn' % (node['key']),
recursive=True).value).encode('ascii', 'ignore')
nodes_dets.append({'node_id': node_id, 'fqdn': fqdn})
return nodes_dets
except EtcdKeyNotFound:
return nodes_dets
except EtcdConnectionFailed as ex:
raise TendrlPerformanceMonitoringException(str(ex))
def initiate_config_generation(node_det):
try:
plugin = NodeMonitoringPlugin(
plugin_name=node_det['plugin'],
node_id=node_det.get('node_id')
)
if plugin.exists():
# More powers like fixed retrials can be added here.This is common
# point through which all monitoring plugin configuration jobs land
# into etcd and hence any action here is reflected to all of them.
return
job_params = {
'node_ids': [node_det.get('node_id')],
"run": 'node_monitoring.flows.ConfigureCollectd',
'type': 'monitoring',
"parameters": {
'plugin_name': node_det['plugin'],
'plugin_conf_params': json.dumps(
node_det['plugin_conf']
).encode('utf-8'),
'Node.fqdn': node_det['fqdn'],
'Service.name': 'collectd',
},
}
job_id = str(uuid.uuid4())
Job(
job_id=job_id,
status='new',
payload=job_params,
).save()
NodeMonitoringPlugin(
plugin_name=node_det['plugin'],
node_id=node_det.get('node_id'),
job_id=job_id
).save(update=False)
except (EtcdException, EtcdConnectionFailed, Exception) as ex:
raise TendrlPerformanceMonitoringException(
'Failed to intiate monitoring configuration for plugin \
%s on %s with parameters %s.Error %s' % (
node_det['plugin'],
node_det['fqdn'],
json.dumps(node_det['plugin_conf']),
str(ex)
)
)
def get_configs():
# TODO(Anmol) : Attempt reading:
# /_tendrl/config/performance_monitoring/clusters/{cluster-id} and if
# not already present, default back to defaults in:
# /_tendrl/config/performance_monitoring
try:
configs = ''
conf = NS._int.client.read('_NS/performance_monitoring/config')
configs = conf.value
return yaml.safe_load(configs)
except (EtcdKeyNotFound, EtcdConnectionFailed, ValueError, SyntaxError,
EtcdException) as ex:
Event(
ExceptionMessage(priority="debug",
publisher=NS.publisher_id,
payload={
"message":
'Fetching monitoring configurations failed.',
"exception": ex
}))
raise TendrlPerformanceMonitoringException(str(ex))
def get_timeseriesnamefromresource(self, **kwargs):
# If in future this function starts to appear more plugin
# specific move it from here to respecive TimeSeriesDBPlugin
delimeter = self.get_plugin().get_delimeter()
resource_name = kwargs['resource_name']
if 'utilization_type' in kwargs:
kwargs['utilization_type'] = self.get_plugin().get_utilizationtype(
resource_name,
kwargs['utilization_type']
)
pattern = {
pm_consts.SYSTEM_UTILIZATION: '$sds_type{0}utilization{0}'
'$utilization_type',
pm_consts.CLUSTER_UTILIZATION: 'cluster_$cluster_id{0}'
'cluster_utilization{0}$utilization_type',
pm_consts.CLUSTER_THROUGHPUT: 'cluster_$cluster_id{0}'
'throughput{0}$network_type{0}$utilization_type',
pm_consts.SYSTEM_THROUGHPUT: '$sds_type{0}'
'throughput{0}$network_type{0}$utilization_type',
pm_consts.NODE_THROUGHPUT: '$node_name{0}'
'network_throughput-$network_type{0}$utilization_type',
pm_consts.LATENCY: 'ping{0}ping-$underscored_monitoring_node_name',
pm_consts.IOPS: 'cluster_$cluster_id{0}cluster_iops_read_write{0}'
'$utilization_type',
pm_consts.SWAP: 'swap{0}$utilization_type',
pm_consts.SWAP_TOTAL: '$utilization_type',
pm_consts.CPU: '$underscored_node_name{0}cpu{0}cpu_system_user{0}'
'$utilization_type',
pm_consts.STORAGE: '$underscored_node_name{0}storage{0}'
'$utilization_type',
pm_consts.CLUSTER_IOPS: 'cluster_$cluster_id{0}'
'cluster_iops_read_write{0}gauge-total'
}
if not pattern.get(resource_name):
raise TendrlPerformanceMonitoringException(
'No pattern found for the requested resource %s.'
)
return Template(
pattern.get(resource_name).format(delimeter)
).substitute(kwargs)
def get_metrics(self, entity_name):
url = 'http://%s:%s/metrics/index.json' % (self.host, str(self.port))
try:
time.sleep(5)
resp = self.http.request('GET', url, timeout=5)
if resp.status != 200:
raise TendrlPerformanceMonitoringException(
'Request status code: %s' % str(resp.status_code)
)
data = resp.data
metrics = ast.literal_eval(data)
result = []
prefix = "%s.%s." % (self.prefix, entity_name.replace('.', '_'))
split_metrics = []
for metric in metrics:
if metric.startswith(prefix):
split_metrics = metric.split(prefix)
result.append(split_metrics[1])
return str(result)
except (ValueError, Exception) as ex:
LOG.error('Failed to get metrics for %s.Error %s ' %
(entity_name, ex), exc_info=True)
def initiate_config_generation(self, conf_name, data, node_det):
try:
job = {
'node_ids': [node_det.get('node_id')],
"run":
'tendrl.node_monitoring.flows.configure_collectd.ConfigureCollectd',
'status': 'new',
'type': 'monitoring',
'integration_id': tendrl_ns.tendrl_context.integration_id,
"parameters": {
'plugin_name': conf_name,
'plugin_conf_params': json.dumps(data),
'Node.fqdn': node_det['fqdn'],
'Service.name': 'collectd',
},
}
tendrl_ns.etcd_orm.client.write("/queue/%s" % str(uuid.uuid4()),
json.dumps(job))
except (EtcdException, EtcdConnectionFailed, EtcdException) as ex:
LOG.error('Failed to intiate monitoring configuration for plugin \
%s on %s with parameters %s.Error %s' %
(conf_name, node_det['fqdn'], data, ex))
raise TendrlPerformanceMonitoringException(str(ex))