Load Balancing as a Service (LBaaS) for OpenStack
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 
 
 

269 lines
9.5 KiB

#! /usr/bin/env python
# Copyright 2014 Hewlett-Packard Development Company, L.P.
#
# 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 errno
import os
import queue
import stat
import time
from oslo_config import cfg
from oslo_log import log as logging
import simplejson
from octavia.amphorae.backends.agent.api_server import util
from octavia.amphorae.backends.health_daemon import health_sender
from octavia.amphorae.backends.utils import haproxy_query
from octavia.amphorae.backends.utils import keepalivedlvs_query
CONF = cfg.CONF
LOG = logging.getLogger(__name__)
SEQ = 0
# MSG_VER is an incrementing integer heartbeat message format version
# this allows for backward compatibility when the amphora-agent is older
# than the controller version and the message format has backwards
# incompatible changes.
#
# ver 1 - Adds UDP listener status when no pool or members are present
# ver 2 - Switch to all listeners in a single combined haproxy config
# ver 3 - Switch stats reporting to deltas
MSG_VER = 3
DELTA_METRICS = ('bin', 'bout', 'ereq', 'stot')
# Filesystem persistent counters for statistics deltas
COUNTERS = None
COUNTERS_FILE = None
def get_counters_file():
global COUNTERS_FILE
if COUNTERS_FILE is None:
stats_file_path = os.path.join(
CONF.haproxy_amphora.base_path, "stats_counters.json")
# Open for read+write and create if necessary
flags = os.O_RDWR | os.O_CREAT
# mode 00644
mode = stat.S_IRUSR | stat.S_IWUSR | stat.S_IRGRP
try:
COUNTERS_FILE = os.fdopen(
os.open(stats_file_path, flags, mode), 'r+')
except OSError:
LOG.info("Failed to open `%s`, ignoring...", stats_file_path)
COUNTERS_FILE.seek(0)
return COUNTERS_FILE
def get_counters():
global COUNTERS
if COUNTERS is None:
try:
COUNTERS = simplejson.load(get_counters_file()) or {}
except (simplejson.JSONDecodeError, AttributeError):
COUNTERS = {}
return COUNTERS
def persist_counters():
"""Attempt to persist the latest statistics values"""
if COUNTERS is None:
return
try:
stats = simplejson.dumps(COUNTERS)
counters_file = get_counters_file()
counters_file.truncate(0)
counters_file.write(stats)
counters_file.flush()
except (OSError, AttributeError):
LOG.warning("Couldn't persist statistics counter file!")
def list_sock_stat_files(hadir=None):
stat_sock_files = {}
if hadir is None:
hadir = CONF.haproxy_amphora.base_path
lb_ids = util.get_loadbalancers()
for lb_id in lb_ids:
sock_file = lb_id + ".sock"
stat_sock_files[lb_id] = os.path.join(hadir, sock_file)
return stat_sock_files
def run_sender(cmd_queue):
LOG.info('Health Manager Sender starting.')
sender = health_sender.UDPStatusSender()
keepalived_cfg_path = util.keepalived_cfg_path()
keepalived_pid_path = util.keepalived_pid_path()
while True:
try:
# If the keepalived config file is present check
# that it is running, otherwise don't send the health
# heartbeat
if os.path.isfile(keepalived_cfg_path):
# Is there a pid file for keepalived?
with open(keepalived_pid_path, 'r') as pid_file:
pid = int(pid_file.readline())
os.kill(pid, 0)
message = build_stats_message()
sender.dosend(message)
except (IOError, OSError) as e:
if e.errno == errno.ENOENT:
# Missing PID file, skip health heartbeat.
LOG.error('Missing keepalived PID file %s, skipping health '
'heartbeat.', keepalived_pid_path)
elif e.errno == errno.ESRCH:
# Keepalived is not running, skip health heartbeat.
LOG.error('Keepalived is configured but not running, '
'skipping health heartbeat.')
else:
LOG.exception('Failed to check keepalived and haproxy status '
'due to exception %s, skipping health '
'heartbeat.', e)
except Exception as e:
LOG.exception('Failed to check keepalived and haproxy status due '
'to exception %s, skipping health heartbeat.', e)
try:
cmd = cmd_queue.get_nowait()
if cmd == 'reload':
LOG.info('Reloading configuration')
CONF.reload_config_files()
elif cmd == 'shutdown':
LOG.info('Health Manager Sender shutting down.')
break
except queue.Empty:
pass
time.sleep(CONF.health_manager.heartbeat_interval)
def get_stats(stat_sock_file):
stats_query = haproxy_query.HAProxyQuery(stat_sock_file)
stats = stats_query.show_stat()
pool_status = stats_query.get_pool_status()
return stats, pool_status
def calculate_stats_deltas(listener_id, row):
counters = get_counters()
listener_counters = counters.get(listener_id, {})
counters[listener_id] = listener_counters
delta_values = {}
for metric_key in DELTA_METRICS:
current_value = int(row[metric_key])
# Get existing counter for our metrics
last_value = listener_counters.get(metric_key, 0)
# Store the new absolute value
listener_counters[metric_key] = current_value
# Calculate a delta for each metric
delta = current_value - last_value
# Did HAProxy restart or reset counters?
if delta < 0:
delta = current_value # If so, reset ours.
delta_values[metric_key] = delta
return delta_values
def build_stats_message():
"""Build a stats message based on retrieved listener statistics.
Example version 3 message without UDP (note that values are deltas,
not absolutes)::
{"id": "<amphora_id>",
"seq": 67,
"listeners": {
"<listener_id>": {
"status": "OPEN",
"stats": {
"tx": 0,
"rx": 0,
"conns": 0,
"totconns": 0,
"ereq": 0
}
}
},
"pools": {
"<pool_id>:<listener_id>": {
"status": "UP",
"members": {
"<member_id>": "no check"
}
}
},
"ver": 3
}
"""
global SEQ
msg = {'id': CONF.amphora_agent.amphora_id,
'seq': SEQ, 'listeners': {}, 'pools': {},
'ver': MSG_VER}
SEQ += 1
stat_sock_files = list_sock_stat_files()
# TODO(rm_work) There should only be one of these in the new config system
for lb_id, stat_sock_file in stat_sock_files.items():
if util.is_lb_running(lb_id):
(stats, pool_status) = get_stats(stat_sock_file)
for row in stats:
if row['svname'] == 'FRONTEND':
listener_id = row['pxname']
delta_values = calculate_stats_deltas(listener_id, row)
msg['listeners'][listener_id] = {
'status': row['status'],
'stats': {'tx': delta_values['bout'],
'rx': delta_values['bin'],
'conns': int(row['scur']),
'totconns': delta_values['stot'],
'ereq': delta_values['ereq']}}
for pool_id, pool in pool_status.items():
msg['pools'][pool_id] = {"status": pool['status'],
"members": pool['members']}
# UDP listener part
udp_listener_ids = util.get_udp_listeners()
if udp_listener_ids:
listeners_stats = keepalivedlvs_query.get_udp_listeners_stats()
if listeners_stats:
for listener_id, listener_stats in listeners_stats.items():
delta_values = calculate_stats_deltas(
listener_id, listener_stats['stats'])
pool_status = keepalivedlvs_query.get_udp_listener_pool_status(
listener_id)
udp_listener_dict = dict()
udp_listener_dict['status'] = listener_stats['status']
udp_listener_dict['stats'] = {
'tx': delta_values['bout'],
'rx': delta_values['bin'],
'conns': listener_stats['stats']['scur'],
'totconns': delta_values['stot'],
'ereq': delta_values['ereq']
}
if pool_status:
pool_id = pool_status['lvs']['uuid']
msg['pools'][pool_id] = {
"status": pool_status['lvs']['status'],
"members": pool_status['lvs']['members']
}
msg['listeners'][listener_id] = udp_listener_dict
persist_counters()
return msg