关于flume的几道题
1，要求：监听一个tcp,udp端口41414将数据打印在控制台
# example.conf: A single-node Flume configuration # Name the components on this agent
a1.sources = r1
a1.sinks = k1
a1.channels = c1 # Describe/configure the source
a1.sources.r1.type = netcat
a1.sources.r1.bind = 0.0.0.0
a1.sources.r1.port = # 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 =
a1.channels.c1.transactionCapacity = # Bind the source and sink to the channel
a1.sources.r1.channels = c1
a1.sinks.k1.channel = c1
启动命令：
bin/flume-ng agent --conf conf/ --conf-file conf/one.conf --name a1 -Dflume.root.logger=INFO,console &
Telnet：
root@Ubuntu-:~# telnet 0.0.0.0
Trying 0.0.0.0...
Connected to 0.0.0.0.
Escape character is '^]'.
huxing
OK
结果：
2，要求：将A机器的日志文件access.log传输到机器B上，并打印到控制台上
这里我假设A机器是131，B机器是132，则 需要将配置文件写在132上，然后正常启动132，而131中只需要启动avro_client，通过avro序列化将文件打到132中。
132中的配置文件内容：
# example.conf: A single-node Flume configuration # 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.bind = 0.0.0.0
a1.sources.r1.port = # 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 =
a1.channels.c1.transactionCapacity = # Bind the source and sink to the channel
a1.sources.r1.channels = c1
a1.sinks.k1.channel = c1
启动132的flume：
bin/flume-ng agent --conf conf/ --conf-file conf/two.conf --name a1 -Dflume.root.logger=INFO,console &
启动131的avro_client：
bin/flume-ng avro-client --host 192.168.22.132 --port --filename logs/avro.log
查看132控制台：
成功
3，监听一个日志文件access.log，如果有日志追加及时的将数据打印在控制台上，如果是大文件呢？堆？
conf内容：
# example.conf: A single-node Flume configuration # Name the components on this agent
a1.sources = r1
a1.sinks = k1
a1.channels = c1 # Describe/configure the source
a1.sources.r1.type = exec
a1.sources.r1.command = tail -F /opt/logs/access.log # 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 =
a1.channels.c1.transactionCapacity = # Bind the source and sink to the channel
a1.sources.r1.channels = c1
a1.sinks.k1.channel = c1
启动命令：
bin/flume-ng agent --conf conf/ --conf-file conf/three.conf --name a1 -Dflume.root.logger=INFO,console &
打文件到控制台：
root@Ubuntu-:/usr/local/apache-flume/logs# cat hu.log >> avro.log
成功
----------------------------------------------------------------------------------
如果是个很大文件的话怎么办呢？
--将这个文件
中的
的注释消掉。
4，A,B机器中的access.log汇总到C机器上然后统一收集到hdfs上分天存储。
在132,135中写入four_avro_sink.conf文件：
# example.conf: A single-node Flume configuration # Name the components on this agent
a1.sources = r1
a1.sinks = k1
a1.channels = c1 # Describe/configure the source
a1.sources.r1.type = exec
a1.sources.r1.command = tail -F /usr/local/apache-flume/logs/avro.log # Describe the sink
a1.sinks.k1.type = avro
a1.sinks.k1.hostname = 192.168.22.131
a1.sinks.k1.port = # Use a channel which buffers events in memory
a1.channels.c1.type = memory
a1.channels.c1.capacity =
a1.channels.c1.transactionCapacity = # Bind the source and sink to the channel
a1.sources.r1.channels = c1
a1.sinks.k1.channel = c1
就是将以exec形式持续的输出最新的数据到sink,再以avro的方式将文件序列化的方式传到131的sink上
启动flume：
root@Ubuntu-:/usr/local/apache-flume# bin/flume-ng agent --conf conf/ --conf-file conf/four_avro_sink.conf --name a1 -Dflume.root.logger=INFO,console &
在131中写入four.conf文件：
#定义agent名， source、channel、sink的名称
access.sources = r1
access.channels = c1
access.sinks = k1 #具体定义source
access.sources.r1.type = avro
access.sources.r1.bind = 0.0.0.0
access.sources.r1.port = #具体定义channel
access.channels.c1.type = memory
access.channels.c1.capacity =
access.channels.c1.transactionCapacity = #定义拦截器，为消息添加时间戳
access.sources.r1.interceptors = i1
access.sources.r1.interceptors.i1.type = org.apache.flume.interceptor.TimestampInterceptor$Builder #具体定义sink
access.sinks.k1.type = hdfs
access.sinks.k1.hdfs.path = hdfs://Ubuntu-1:9000/%Y%m%d
access.sinks.k1.hdfs.filePrefix = events-
access.sinks.k1.hdfs.fileType = DataStream
#access.sinks.k1.hdfs.fileType = CompressedStream
#access.sinks.k1.hdfs.codeC = gzip
#不按照条数生成文件
access.sinks.k1.hdfs.rollCount =
#HDFS上的文件达到64M时生成一个文件
access.sinks.k1.hdfs.rollSize =
access.sinks.k1.hdfs.rollInterval = #组装source、channel、sink
access.sources.r1.channels = c1
access.sinks.k1.channel = c1
启动Hadoop：
root@Ubuntu-:/usr/local/hadoop-2.6.# sbin/start-dfs.sh
启动flume：
root@Ubuntu-:/usr/local/apache-flume# bin/flume-ng agent --conf conf/ --conf-file conf/four.conf --name access -Dflume.root.logger=INFO,console &
5，A，B，机器中的access.log ugcheader.log ugctail.log汇总到C机器上。然后统一收集到HDFS的不同目录上
改成
access.sinks.k1.hdfs.path = hdfs://Ubuntu-1:9000/%{type}/%Y%m%d
另132中的配置文件：
# example.conf: A single-node Flume configuration # Name the components on this agent
a1.sources = r1 r2 r3
a1.sinks = k1
a1.channels = c1 # Describe/configure the source
a1.sources.r1.type = exec
a1.sources.r1.command = tail -F /usr/local/apache-flume/logs/avro.log
a1.sources.r1.interceptors = i1
a1.sources.r1.interceptors.i1.type = static
a1.sources.r1.interceptors.i1.key = type
a1.sources.r1.interceptors.i1.value = access a1.sources.r2.type = exec
a1.sources.r2.command = tail -F /usr/local/apache-flume/logs/flume.log
a1.sources.r2.interceptors = i2
a1.sources.r2.interceptors.i2.type = static
a1.sources.r2.interceptors.i2.key = type
a1.sources.r2.interceptors.i2.value = ugchead a1.sources.r3.type = exec
a1.sources.r3.command = tail -F /usr/local/apache-flume/logs/hu.log
a1.sources.r3.interceptors = i3
a1.sources.r3.interceptors.i3.type = static
a1.sources.r3.interceptors.i3.key = type
a1.sources.r3.interceptors.i3.value = ugctail # Describe the sink
a1.sinks.k1.type = avro
a1.sinks.k1.hostname = 192.168.22.131
a1.sinks.k1.port = #a1.sinks.k1.type = logger # Use a channel which buffers events in memory
a1.channels.c1.type = memory
a1.channels.c1.capacity =
a1.channels.c1.transactionCapacity = # Bind the source and sink to the channel
a1.sources.r1.channels = c1
a1.sources.r2.channels = c1
a1.sources.r3.channels = c1
a1.sinks.k1.channel = c1
6，access.log收集后指定多个目的地【同时，打印到控制台、输出到HDFS】
131中：
#定义agent名， source、channel、sink的名称
access.sources = r1
access.channels = c1 c2
access.sinks = k1 k2 #具体定义source
access.sources.r1.type = avro
access.sources.r1.bind = 0.0.0.0
access.sources.r1.port = #具体定义channel
access.channels.c1.type = memory
access.channels.c1.capacity =
access.channels.c1.transactionCapacity = access.channels.c2.type = memory
access.channels.c2.capacity =
access.channels.c2.transactionCapacity = access.sinks.k2.type = logger ！！！！重点是这里的k2！！！！！ #定义拦截器，为消息添加时间戳
access.sources.r1.interceptors = i1
access.sources.r1.interceptors.i1.type = org.apache.flume.interceptor.TimestampInterceptor$Builder #具体定义sink
access.sinks.k1.type = hdfs
access.sinks.k1.hdfs.path = hdfs://Ubuntu-1:9000/source/%{type}/%Y%m%d
access.sinks.k1.hdfs.filePrefix = events-
access.sinks.k1.hdfs.fileType = DataStream
#access.sinks.k1.hdfs.fileType = CompressedStream
#access.sinks.k1.hdfs.codeC = gzip
#不按照条数生成文件
access.sinks.k1.hdfs.rollCount =
#HDFS上的文件达到64M时生成一个文件
access.sinks.k1.hdfs.rollSize =
access.sinks.k1.hdfs.rollInterval = #组装source、channel、sink access.sources.r1.channels = c1 c2
access.sinks.k1.channel = c1
access.sinks.k2.channel = c2
132中还是之前第5题中的配置
7，在程序里打印日志到flume根据不同的业务指定不同的目的地【控制台、avro】，查看日志的log4j日志的header
pom文件：
<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
<modelVersion>4.0.</modelVersion> <groupId>cn.hx</groupId>
<artifactId>FlumeSource</artifactId>
<version>1.0-SNAPSHOT</version>
<packaging>jar</packaging> <name>FlumeSource</name>
<url>http://maven.apache.org</url> <properties>
<project.build.sourceEncoding>UTF-</project.build.sourceEncoding>
<maven.compiler.source>1.8</maven.compiler.source>
<maven.compiler.target>1.8</maven.compiler.target>
</properties> <build>
<pluginManagement>
<plugins>
<plugin>
<groupId>org.apache.maven.plugins</groupId>
<artifactId>maven-jar-plugin</artifactId>
<configuration> <archive>
<manifest>
<mainClass>cn.hx.test</mainClass>
<addClasspath>true</addClasspath>
<classpathPrefix>lib/</classpathPrefix>
</manifest> </archive>
<classesDirectory>
</classesDirectory>
</configuration>
</plugin>
</plugins>
</pluginManagement>
</build> <dependencies>
<dependency>
<groupId>junit</groupId>
<artifactId>junit</artifactId>
<version>3.8.</version>
<scope>test</scope>
</dependency>
<dependency>
<groupId>org.apache.hadoop</groupId>
<artifactId>hadoop-common</artifactId>
<version>2.6.</version>
</dependency>
<dependency>
<groupId>org.apache.hadoop</groupId>
<artifactId>hadoop-client</artifactId>
<version>2.6.</version>
</dependency>
<dependency>
<groupId>org.apache.hadoop</groupId>
<artifactId>hadoop-hdfs</artifactId>
<version>2.6.</version>
</dependency>
<dependency>
<groupId>log4j</groupId>
<artifactId>log4j</artifactId>
<version>1.2.</version>
</dependency>
<dependency>
<groupId>log4j</groupId>
<artifactId>log4j</artifactId>
<version>1.2.</version>
</dependency>
</dependencies>
</project>
loj4j文件：
##<!-- ========================== 自定义输出格式说明================================ -->
##<!-- %p 输出优先级，即DEBUG，INFO，WARN，ERROR，FATAL -->
##<!-- %r 输出自应用启动到输出该log信息耗费的毫秒数 -->
##<!-- %c 输出所属的类目，通常就是所在类的全名 -->
##<!-- %t 输出产生该日志事件的线程名 -->
##<!-- %n 输出一个回车换行符，Windows平台为“/r/n”，Unix平台为“/n” -->
##<!-- %d 输出日志时间点的日期或时间，默认格式为ISO8601，也可以在其后指定格式，比如：%d{yyy MMM dd ##HH:mm:ss,SSS}，输出类似：2002年10月18日 ：：， -->
##<!-- %l 输出日志事件的发生位置，包括类目名、发生的线程，以及在代码中的行数。举例：Testlog4.main(TestLog4.java:) -->
##<!-- ========================================================================== --> ### set log levels ### #默认logger
#INFO是指级别不小于INFO的日志才会使用stdoutappender。ERROR、WARN、INFO
log4j.rootLogger=INFO,stdout1 #自定义logger #log4j.logger.accessLogger=INFO,flume
#log4j.logger.ugcLogger=INFO,flume log4j.logger.std1Logger=INFO,stdout1,
log4j.logger.std2Logger=INFO,stdout2 log4j.logger.access=INFO,flume log4j.logger.ugchead=INFO,flume
log4j.logger.ugctail=INFO,flume #某个包的level的appender
#log4j.logger.com.zenith.flume = INFO,flume ### flume ###
log4j.appender.flume=org.apache.flume.clients.log4jappender.Log4jAppender
log4j.appender.flume.layout=org.apache.log4j.PatternLayout
log4j.appender.flume.layout.ConversionPattern=%d{yyyy-MM-dd HH:mm:ss} %c{} [%p] %m%n
log4j.appender.flume.Hostname=192.168.22.131
log4j.appender.flume.Port=
log4j.appender.flume.UnsafeMode = true ### stdout ###
log4j.appender.stdout1=org.apache.log4j.ConsoleAppender
log4j.appender.stdout1.Threshold=DEBUG
log4j.appender.stdout1.Target=System.out
log4j.appender.stdout1.layout=org.apache.log4j.PatternLayout
log4j.appender.stdout1.layout.ConversionPattern=%d{yyyy-MM-dd HH:mm:ss} %c{} [%p] %m%n ### stdout ###
log4j.appender.stdout2=org.apache.log4j.ConsoleAppender
log4j.appender.stdout2.Threshold=DEBUG
log4j.appender.stdout2.Target=System.out
log4j.appender.stdout2.layout=org.apache.log4j.PatternLayout
log4j.appender.stdout2.layout.ConversionPattern=%d{yyyy-MM-dd hh:mm:ss} %c{} [%p] %m%n ### access ###
log4j.appender.access=org.apache.log4j.DailyRollingFileAppender
log4j.appender.access.Threshold=INFO
log4j.appender.access.File=/usr/local/apache-flume/logs/avro.log
log4j.appender.access.Append=true
log4j.appender.access.DatePattern='.'yyyy-MM-dd
log4j.appender.access.layout=org.apache.log4j.PatternLayout
log4j.appender.access.layout.ConversionPattern=%m%n ### ugchead ###
log4j.appender.ugchead=org.apache.log4j.DailyRollingFileAppender
log4j.appender.ugchead.Threshold=INFO
log4j.appender.ugchead.File=/usr/local/apache-flume/logs/flume.log
log4j.appender.ugchead.Append=true
log4j.appender.ugchead.DatePattern='.'yyyy-MM-dd
log4j.appender.ugchead.layout=org.apache.log4j.PatternLayout
log4j.appender.ugchead.layout.ConversionPattern=%m%n ### ugctail ###
log4j.appender.ugctail=org.apache.log4j.DailyRollingFileAppender
log4j.appender.ugctail.Threshold=INFO
log4j.appender.ugctail.File=/usr/local/apache-flume/logs/hu.log
log4j.appender.ugctail.Append=true
log4j.appender.ugctail.DatePattern='.'yyyy-MM-dd
log4j.appender.ugctail.layout=org.apache.log4j.PatternLayout
log4j.appender.ugctail.layout.ConversionPattern=%m%n
程序：
package cn.hx; import org.apache.log4j.BasicConfigurator;
import org.apache.log4j.Logger; /**
* Created by hushiwei on 2017/8/20.
*/
public class test {
protected static final Logger loggeaccess = Logger.getLogger("access"); protected static final Logger loggerugc = Logger.getLogger("ugchead"); public static void main(String[] args) throws Exception {
BasicConfigurator.configure(); while (true) {
loggeaccess.info("this is acccess log");
loggerugc.info("ugc");
//KafkaUtil util=new KafkaUtil();
//util.initProducer();
//util.produceData("crxy","time",String.valueOf(new Date().getTime()));
Thread.sleep();
}
}
}
在131中执行：
root@Ubuntu-:/usr/local/apache-flume# bin/flume-ng agent --conf conf/ --conf-file conf/avro_source.conf --name agent1 -Dflume.root.logger=INFO,console &
avro.source文件是上面某道题中的文件
打jar包后到131中执行
可是报错，没有解决：
Exception in thread "main" java.lang.NoClassDefFoundError: org/apache/log4j/Logger
Caused by:java.lang.ClassNotFoundException:org.apache.log4j.Logger
8，A机器的access.log日志采集后打印到B、C做负载均衡，打印到控制台上，load_balance
132和135中：
conf文件用avro_source.conf
启动：
root@Ubuntu-:/usr/local/apache-flume# bin/flume-ng agent --conf conf/ --conf-file conf/avro_source.conf --name agent1 -Dflume.root.logger=INFO,console &
131中：
# Name the components on this agent
a1.sources = r1
a1.sinks = k1 k2
a1.channels = c1 # Describe/configure the source
a1.sources.r1.type = exec
a1.sources.r1.channels=c1
a1.sources.r1.command=tail -F /usr/local/apache-flume/logs/xing.log #define sinkgroups
a1.sinkgroups=g1
a1.sinkgroups.g1.sinks=k1 k2
a1.sinkgroups.g1.processor.type=load_balance
a1.sinkgroups.g1.processor.backoff=true
a1.sinkgroups.g1.processor.selector=round_robin #define the sink 1
a1.sinks.k1.type=avro
a1.sinks.k1.hostname=192.168.22.132
a1.sinks.k1.port= #define the sink 2
a1.sinks.k2.type=avro
a1.sinks.k2.hostname=192.168.22.135
a1.sinks.k2.port= # Use a channel which buffers events in memory
a1.channels.c1.type = memory
a1.channels.c1.capacity =
a1.channels.c1.transactionCapacity = # Bind the source and sink to the channel
a1.sources.r1.channels = c1
a1.sinks.k1.channel = c1
a1.sinks.k2.channel=c1
启动
root@Ubuntu-:/usr/local/apache-flume# bin/flume-ng agent --conf conf/ --conf-file conf/eight.conf --name a1 -Dflume.root.logger=INFO,console &
在131中：
在132中：
在135中：
9，A机器的access.log日志采集后打印到B、C做故障转移，打印到控制台上，failover
132和135中起avro_source的conf文件
131中启：
# Name the components on this agent
a1.sources = r1
a1.sinks = k1 k2
a1.channels = c1 # Describe/configure the source
a1.sources.r1.type = exec
a1.sources.r1.channels=c1
a1.sources.r1.command=tail -F /usr/local/apache-flume/logs/xing.log #define sinkgroups
a1.sinkgroups=g1
a1.sinkgroups.g1.sinks=k1 k2
a1.sinkgroups.g1.processor.type=failover
a1.sinkgroups.g1.processor.priority.k1=
a1.sinkgroups.g1.processor.priority.k2=
a1.sinkgroups.g1.processor.maxpenalty= #define the sink 1
a1.sinks.k1.type=avro
a1.sinks.k1.hostname=192.168.22.132
a1.sinks.k1.port= #define the sink 2
a1.sinks.k2.type=avro
a1.sinks.k2.hostname=192.168.22.135
a1.sinks.k2.port= # Use a channel which buffers events in memory
a1.channels.c1.type = memory
a1.channels.c1.capacity =
a1.channels.c1.transactionCapacity = # Bind the source and sink to the channel
a1.sources.r1.channels = c1
a1.sinks.k1.channel = c1
a1.sinks.k2.channel=c1
启131
root@Ubuntu-:/usr/local/apache-flume# bin/flume-ng agent --conf conf/ --conf-file conf/nine.conf --name a1 -Dflume.root.logger=INFO,console &
查看：
关闭132中的flume之后
132宕机之后 可以看到数据直接转到135中了：
关于flume的几道题的更多相关文章
- Flume1 初识Flume和虚拟机搭建Flume环境
前言: 工作中需要同步日志到hdfs,以前是找运维用rsync做同步,现在一般是用flume同步数据到hdfs.以前为了工作简单看个flume的一些东西,今天下午有时间自己利用虚拟机搭建了 ...
- 处理Assetbundle依赖关系时想到的一道题
在处理unit3d的assetbundle依赖关系的时候,想到了一道有趣的题目: 给定一堆数据,例如{A = {1, 3, 4}, B = {3, 4}, C = {5, 6}, D = {6, 7, ...
- Flume（4）实用环境搭建：source(spooldir)+channel(file)+sink(hdfs)方式
一.概述: 在实际的生产环境中,一般都会遇到将web服务器比如tomcat.Apache等中产生的日志倒入到HDFS中供分析使用的需求.这里的配置方式就是实现上述需求. 二.配置文件: #agent1 ...
- Flume（3）source组件之NetcatSource使用介绍
一.概述: 本节首先提供一个基于netcat的source+channel(memory)+sink(logger)的数据传输过程.然后剖析一下NetcatSource中的代码执行逻辑. 二.flum ...
- Flume（2）组件概述与列表
上一节搭建了flume的简单运行环境,并提供了一个基于netcat的演示.这一节继续对flume的整个流程进行进一步的说明. 一.flume的基本架构图: 下面这个图基本说明了flume的作用,以及f ...
- Flume（1）使用入门
一.概述: Flume是Cloudera提供的一个高可用的,高可靠的,分布式的海量日志采集.聚合和传输的系统. 当前Flume有两个版本Flume 0.9X版本的统称Flume-og,Flume1.X ...
- 大数据平台架构（flume＋kafka＋hbase＋ELK+storm+redis+mysql）
上次实现了flume+kafka+hbase+ELK:http://www.cnblogs.com/super-d2/p/5486739.html 这次我们可以加上storm: storm-0.9.5 ...
- flume+kafka+spark streaming整合
1.安装好flume2.安装好kafka3.安装好spark4.流程说明: 日志文件->flume->kafka->spark streaming flume输入:文件 flume输 ...
- flume使用示例
flume的特点: flume是一个分布式.可靠.和高可用的海量日志采集.聚合和传输的系统.支持在日志系统中定制各类数据发送方,用于收集数据;同时,Flume提供对数据进行简单处理,并写到各种数据接受 ...
随机推荐
- Question | 关于Android安全的一二事
本文来自网易云社区 "Question"为网易云易盾的问答栏目,将会解答和呈现安全领域大家常见的问题和困惑.如果你有什么疑惑,也欢迎通过邮件(zhangyong02@corp.ne ...
- NoSQL简单学习（一）
只是简单的知道有这个东西,却从来没有去接触,今天看了几篇文章,记录一下,开始慢慢接触这一领域 简介: 8种Nosql数据库系统对比 http://blog.jobbole.com/1344/ 一网打尽 ...
- LightGBM详细用法--机器学习算法--周振洋
LightGBM算法总结 2018年08月21日 18:39:47 Ghost_Hzp 阅读数:2360 版权声明:本文为博主原创文章,未经博主允许不得转载. https://blog.csdn.ne ...
- 2.爬虫 urlib库讲解 异常处理、URL解析、分析Robots协议
1.异常处理 URLError类来自urllib库的error模块,它继承自OSError类,是error异常模块的基类,由request模块产生的异常都可以通过这个类来处理. from urllib ...
- Markdown常用的几种语法
在VScode上面写的,现将代码粘贴如下:(在VScode里运行下即可) # Markdown语法 # Ctrl + k v 打开侧边预览 ## 一.加粗斜体删除线 **这是要加粗的文字** *这是要 ...
- Python第二天 （数据类型，变量 ）
1. 把任意数据类型赋值给变量 在Python中,等号=是赋值语句,可以把任意数据类型赋值给变量,同一个变量可以反复赋值,而且可以是不同类型的变量,例如: 例子:a = 123 # a是整数 prin ...
- Visual Studio 6.0安装包
点击下载
- 【PHP】- include、require、include_once 和 require_once的区别
1.include:会将指定的档案读入并且执行里面的程序. 被导入的档案中的程序代码都会被执行,而且这些程序在执行的时候会拥有和源文件中呼叫到 include() 函数的位置相同的变量范围( ...
- Session接口常用方法
org.hibernate.Session接口 beginTransaction 开启事务 clear 清缓存 close 关闭session connection - 过时 获取Connection ...
- [洛谷P1231] 教辅的组成
题目大意:有n1本书,n2本练习册和n3个答案,然后又一些条件,说明某本答案可能和某本书对应,某本练习册可能和某本书对应,求最多有多少本完整的书(有书,练习册,答案) 题解:网络流,对应就连边,然后考 ...