Witek Bedyk 0696cf54ee Add configuration option influxdb.batch_size
The new configuration option is used to control the maximum size of
batches written to InfluxDB and optimizes performance for high loads.
The default value is set to 10 000 data points per batch.

Change-Id: I82c3e4e64984a996e41fe2657ce36b032b40d3fd
Story: 2007191
Task: 38319
2020-01-21 11:07:22 +01:00

79 lines
3.5 KiB
Python

# (C) Copyright 2016-2017 Hewlett Packard Enterprise Development LP
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
# implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import abc
import influxdb
from oslo_config import cfg
import six
from monasca_persister.repositories import abstract_repository
from monasca_persister.repositories import data_points
DATABASE_NOT_FOUND_MSG = "database not found"
@six.add_metaclass(abc.ABCMeta)
class AbstractInfluxdbRepository(abstract_repository.AbstractRepository):
def __init__(self):
super(AbstractInfluxdbRepository, self).__init__()
self.conf = cfg.CONF
self._influxdb_client = influxdb.InfluxDBClient(
self.conf.influxdb.ip_address,
self.conf.influxdb.port,
self.conf.influxdb.user,
self.conf.influxdb.password)
if self.conf.influxdb.db_per_tenant:
self.data_points_class = data_points.DataPointsAsDict
else:
self.data_points_class = data_points.DataPointsAsList
def write_batch(self, data_points):
if self.conf.influxdb.db_per_tenant:
for tenant_id, tenant_data_points in data_points.items():
database = '%s_%s' % (self.conf.influxdb.database_name,
tenant_id)
self._write_batch(tenant_data_points, database)
else:
self._write_batch(data_points, self.conf.influxdb.database_name)
def _write_batch(self, data_points, database):
# NOTE (brtknr): Loop twice to ensure database is created if missing.
for retry in range(2):
try:
batch_size = self.conf.influxdb.batch_size
self._influxdb_client.write_points(data_points, 'ms',
protocol='line',
database=database,
batch_size=batch_size)
break
except influxdb.exceptions.InfluxDBClientError as ex:
# When a databse is not found, the returned exception resolves
# to: {"error":"database not found: \"test\""}
if DATABASE_NOT_FOUND_MSG in str(ex):
self._influxdb_client.create_database(database)
# NOTE (brtknr): Only apply default retention policy at
# database creation time so that existing policies are
# not overridden since administrators may want different
# retention policy per tenant.
hours = self.conf.influxdb.default_retention_hours
if hours > 0:
rp = '{}h'.format(hours)
default_rp = dict(database=database, default=True,
name=rp, duration=rp,
replication='1')
self._influxdb_client.create_retention_policy(**default_rp)
else:
raise