Apache Flume基础（四）负载均衡与容错机制

负载均衡

负载均衡是用于解决一台机器(一个进程)无法解决所有请求而产生的一种

算法。Load balancing Sink Processor 能够实现 load balance 功能，如下图 Agent1 是一个路由节点，负责将 Channel 暂存的 Event 均衡到对应的多个 Sink 组件上，而每个 Sink 组件分别连接到一个独立的 Agent 上，示例配置，如下所示：

image

第一台机器配置

#agent1 name
agent1.channels = c1
agent1.sources = r1
agent1.sinks = k1 k2

#set gruop
agent1.sinkgroups = g1

#set channel
agent1.channels.c1.type = memory
agent1.channels.c1.capacity = 1000
agent1.channels.c1.transactionCapacity = 100

agent1.sources.r1.channels = c1
agent1.sources.r1.type = exec
#在路径中的哪个文件夹
agent1.sources.r1.command = tail -F /root/logs/123.log

# set sink1 负载一
agent1.sinks.k1.channel = c1
agent1.sinks.k1.type = avro
agent1.sinks.k1.hostname = node-2
agent1.sinks.k1.port = 52020

# set sink2 负载二
agent1.sinks.k2.channel = c1
agent1.sinks.k2.type = avro
agent1.sinks.k2.hostname = node-3
agent1.sinks.k2.port = 52020

#set sink group
agent1.sinkgroups.g1.sinks = k1 k2

#set failover
agent1.sinkgroups.g1.processor.type = load_balance
agent1.sinkgroups.g1.processor.backoff = true
agent1.sinkgroups.g1.processor.selector = round_robin
agent1.sinkgroups.g1.processor.selector.maxTimeOut=10000

负载机一

# Name the components on this agent
a1.sources = r1
a1.sinks = k1
a1.channels = c1

# Describe/configure the source
a1.sources.r1.type = avro
a1.sources.r1.channels = c1

#本机host名称
a1.sources.r1.bind = node-2
a1.sources.r1.port = 52020

# Describe the sink
a1.sinks.k1.type = logger

# Use a channel which buffers events in memory
a1.channels.c1.type = memory
a1.channels.c1.capacity = 1000
a1.channels.c1.transactionCapacity = 100

# Bind the source and sink to the channel
a1.sources.r1.channels = c1
a1.sinks.k1.channel = c1

负载机二

# Name the components on this agent
a1.sources = r1
a1.sinks = k1
a1.channels = c1

# Describe/configure the source
a1.sources.r1.type = avro
a1.sources.r1.channels = c1

#本机host名称
a1.sources.r1.bind = node-3
a1.sources.r1.port = 52020

# Describe the sink
a1.sinks.k1.type = logger

# Use a channel which buffers events in memory
a1.channels.c1.type = memory
a1.channels.c1.capacity = 1000
a1.channels.c1.transactionCapacity = 100

# Bind the source and sink to the channel
a1.sources.r1.channels = c1
a1.sinks.k1.channel = c1

在flume根目录执行

bin/flume-ng agent -c conf -f conf/avro-logger.conf -n a1 -Dflume.root.logger=INFO,console

容错机制

Failover Sink Processor能够实现 failover 功能，具体流程类似 load balance，但是内部处理机制与 load balance 完全不同。

Failover Sink Processor 维护一个优先级 Sink 组件列表，只要有一个 Sink 组件可用，Event 就被传递到下一个组件。故障转移机制的作用是将失败的 Sink 降级到一个池，在这些池中它们被分配一个冷却时间，随着故障的连续，在重试 之前冷却时间增加。一旦 Sink 成功发送一个事件，它将恢复到活动池。 Sink 具 有与之相关的优先级，数量越大，优先级越高。

例如，具有优先级为 100 的 sink 在优先级为 80 的 Sink 之前被激活。如果 在发送事件时汇聚失败，则接下来将尝试下一个具有最高优先级的 Sink 发送事件。如果没有指定优先级，则根据在配置中指定 Sink 的顺序来确定优先级。

主机配置

#agent1 name
agent1.channels = c1
agent1.sources = r1
agent1.sinks = k1 k2

#set gruop
agent1.sinkgroups = g1

#set channel
agent1.channels.c1.type = memory
agent1.channels.c1.capacity = 1000
agent1.channels.c1.transactionCapacity = 100

agent1.sources.r1.channels = c1
agent1.sources.r1.type = exec
agent1.sources.r1.command = tail -F /root/logs/456.log

# set sink1 容错机一
agent1.sinks.k1.channel = c1
agent1.sinks.k1.type = avro
agent1.sinks.k1.hostname = node-2
agent1.sinks.k1.port = 52020

# set sink2 容错机二
agent1.sinks.k2.channel = c1
agent1.sinks.k2.type = avro
agent1.sinks.k2.hostname = node-3
agent1.sinks.k2.port = 52020

#set sink group
agent1.sinkgroups.g1.sinks = k1 k2

#set failover
#设置容错
agent1.sinkgroups.g1.processor.type = failover
agent1.sinkgroups.g1.processor.priority.k1 = 10
agent1.sinkgroups.g1.processor.priority.k2 = 1
agent1.sinkgroups.g1.processor.maxpenalty = 10000

容错机一

# Name the components on this agent
a1.sources = r1
a1.sinks = k1
a1.channels = c1

# Describe/configure the source
a1.sources.r1.type = avro
a1.sources.r1.channels = c1

#容错机host
a1.sources.r1.bind = node-2
a1.sources.r1.port = 52020

# Describe the sink
a1.sinks.k1.type = logger

# Use a channel which buffers events in memory
a1.channels.c1.type = memory
a1.channels.c1.capacity = 1000
a1.channels.c1.transactionCapacity = 100

# Bind the source and sink to the channel
a1.sources.r1.channels = c1
a1.sinks.k1.channel = c1

容错机二

# Name the components on this agent
a1.sources = r1
a1.sinks = k1
a1.channels = c1

# Describe/configure the source
a1.sources.r1.type = avro
a1.sources.r1.channels = c1

#容错机host
a1.sources.r1.bind = node-3
a1.sources.r1.port = 52020

# Describe the sink
a1.sinks.k1.type = logger

# Use a channel which buffers events in memory
a1.channels.c1.type = memory
a1.channels.c1.capacity = 1000
a1.channels.c1.transactionCapacity = 100

# Bind the source and sink to the channel
a1.sources.r1.channels = c1
a1.sinks.k1.channel = c1

最后在flume的根目录执行

bin/flume-ng agent -c conf -f conf/exec-avro.conf -n agent1 -Dflume.root.logger=INFO,console

ps：先执行靠近hdfs的机器再执行数据源的机器