cloudera 1.1.0 简介和安装要求说明

1. 关于cloudera，CDH和CM

cloudera是一家hadoop的商业公司，提供hadoop的商业产品CDH(Cloudera Distribution including Apache Hadoop)。同时cloudera也是apache软件基金会的赞助商。

• (CDH)[https://www.cloudera.com/documentation/enterprise/6/6.0/topics/cdh_intro.html]: cloudera根据apache hadoop及其他开源工具组合的一套cloudera的hadoop发行版, 包含以下组件
  • HIVE
  • IMPALA
  • KUDU
  • SENTRY
  • SPARK
• (CM - cloudera manager)[https://www.cloudera.com/documentation/enterprise/6/6.0/topics/cm_intro_primer.html]: cloudera提供的一个程序，可以从web界面管理CDH集群。

2. CDH6.X安装要求和支持的版本

参考文档： (CDH安装要求说明官方文档)[https://www.cloudera.com/documentation/enterprise/6/release-notes/topics/rg_requirements_supported_versions.html]

1) 操作系统要求

• 软件版本：
  • python 2.7
  • perl
  • python-psycopg2 - 如果使用postgresql的话，依赖这个python包
  • iproute package， centos7系统的话，版本需要iproute-3.10
• 系统版本：
  • centos 7.5, 7.4, 7.3, 7.2
  • centos 6.9 , 6.8
• HDFS支持的底层文件系统格式：
  • ext3, HDFS测试最成熟的文件系统格式
  • ext4, HDFS不支持ext3升级为ext4，推荐直接将磁盘格式化为ext4后使用
  • XFS, centos7的默认存储格式
  • S3, 亚马逊的简单存储服务
• KUDU支持的底层文件系统格式：
  • ext4
  • XFS
• file access time：
  • 编辑/etc/fstab: /dev/sdb1 /data1 ext4 defaults,noatime 0, 增加noatime选项，具体磁盘和挂载点按照实际情况来
  • 使上面的配置生效: mount -o remount /data1, 挂载点按照实际情况来
• 文件系统挂载选项： 磁盘挂载到系统时，会有一个sync选项来启用写同步，但是对于HDFS, YARN, Kafka 和 Kudu来说，在CDH中，大部分的写操作已经做到了多副本备份，所以，写同步是没有必要的，降低性能的，可以禁用掉。例如: /dev/sdb1 /data1 ext4 defaults,noatime,nosync 0
• nproc配置： CM会自动在/etc/security/limits.conf中配置nproc限制，但是这个配置可能会被/etc/security/limits.d/中的文件覆盖掉，请确保这样的事情不会发生，或者你手动设定nproc来保证上线足够高，例如262144或65536

2) 数据库要求

CM和CDH自带了一个内嵌的PostgreSQL数据库，仅用于非生产环境，生产环境下，需要配置使用一个外部使用的数据库。

注意点：

• 使用UTF8编码，mysql和mariadb必须使用utf8编码，而不是uft8mb4；
• 如果使用MySQL5.7，需要安装MySQL-shared-compat 或 MySQL-shared package，因为CM的agent安装时依赖这个包。
• 不支持mysql的基于GTID的replication
• Hue需要操作系统默认版本的MySQL和MariaDB版本。
• MySQL社区版和企业版都支持
• cloudera支持oracle的Exadata和RAC instances当他们作为CDH的后端数据库却没有配置高可用时。 cloudera假定后端的数据库都是单点实例, 不支持数据库的高可用。

当重启进程时，各个服务都会在CM连接的数据库中读取配置，如果配置无法读取，大数据集群将不会正确的启动。所以必须要提前做好数据库的备份，这方面内容可以参照(数据库备份)[https://www.cloudera.com/documentation/enterprise/6/latest/topics/cm_ag_backup_dbs.html#xd_583c10bfdbd326ba–6eed2fb8-14349d04bee–7e98]

数据库支持列表：

• MySQL 5.7
• MariaDB 5.5, 10.0
• PostgreSQL 8.4, 9.2, 9.4
• Oracle 12c

3) JAVA

CDH 6.X 仅支持oracle JDK8(不支持openjdk)，以下是详细列表：

• 1.8u162 Recommended / Latest version tested
• 1.8u131 Recommended
• 1.8u121 Recommended
• 1.8u111 Recommended
• 1.8u102 Recommended
• 1.8u91 Recommended
• 1.8u74 Recommended
• 1.8u31 Minimum required

4) 网络和安全要求

CDH和CM的TLS版本支持表

Component	Role	Name	Port	Version
Cloudera Manager	Cloudera Manager Server		7182	TLS 1.2
Cloudera Manager	Cloudera Manager Server		7183	TLS 1.2
Flume			9099	TLS 1.2
Flume		Avro Source/Sink		TLS 1.2
Flume		Flume HTTP Source/Sink		TLS 1.2
HBase	Master	HBase Master Web UI Port	60010	TLS 1.2
HDFS	NameNode	Secure NameNode Web UI Port	50470	TLS 1.2
HDFS	Secondary NameNode	Secure Secondary NameNode Web UI Port	50495	TLS 1.2
HDFS	HttpFS	REST Port	14000	TLS 1.1, TLS 1.2
Hive	HiveServer2	HiveServer2 Port	10000	TLS 1.2
Hue	Hue Server	Hue HTTP Port	8888	TLS 1.2
Impala	Impala Daemon	Impala Daemon Beeswax Port	21000	TLS 1.2
Impala	Impala Daemon	Impala Daemon HiveServer2 Port	21050	TLS 1.2
Impala	Impala Daemon	Impala Daemon Backend Port	22000	TLS 1.2
Impala	Impala StateStore	StateStore Service Port	24000	TLS 1.2
Impala	Impala Daemon	Impala Daemon HTTP Server Port	25000	TLS 1.2
Impala	Impala StateStore	StateStore HTTP Server Port	25010	TLS 1.2
Impala	Impala Catalog Server	Catalog Server HTTP Server Port	25020	TLS 1.2
Impala	Impala Catalog Server	Catalog Server Service Port	26000	TLS 1.2
Oozie	Oozie Server	Oozie HTTPS Port	11443	TLS 1.1, TLS 1.2
Solr	Solr Server	Solr HTTP Port	8983	TLS 1.1, TLS 1.2
Solr	Solr Server	Solr HTTPS Port	8985	TLS 1.1, TLS 1.2
Spark	History Server		18080	TLS 1.2
YARN	ResourceManager	ResourceManager Web Application HTTP Port	8090	TLS 1.2
YARN	JobHistory Server	MRv1 JobHistory Web Application HTTP Port	19890	TLS 1.2

网络和安全要求

• 网络协议： 仅支持ipv4，需要禁用ipv6
• 集群主机的/etc/hosts文件
  • 所有的主机上都要有一致的/etc/hosts文件
  • hostname中不包含大写字母
  • 不包含重复的ip
• 大多数情况下，在使用CM安装和升级集群时，CM需要拥有对所有主机的SSH权限。一旦集群安装或者升级完毕，需要禁用CM的SSH的访问权限。CM不会保存SSH的认证，当安装和升级过程结束，CM会丢弃所有的SSH认证。
• CM agent需要用root身份运行，来保证必要目录得以创建以及让特定的用户拥有进程和文件的权限（例如hdfs和mapred）。
• selinux尽量关闭，或者你有本事别让它阻碍CDH和CM正常运行
• 防火墙需要关闭，或者放行所有CM CDH和相关服务的(端口)[https://www.cloudera.com/documentation/enterprise/6/latest/topics/cm_ig_ports.html#concept_k5z_vwy_4j]
• 对RHEL和CENTOS来说，/etc/sysconfig/network文件必须配置正确的hostname
• CM和CDH使用多个users和groups来完成它们的工作。具体是哪些用户取决于你安装哪些组件和服务，不要去删除这些用户和组，也不要去修改他们的权限。CM、CDH和被管理的服务使用以下用户和组：

Component (Version)	Unix User ID	Groups	Functionality
Cloudera Manager (all versions)	cloudera-scm	cloudera-scm	Clusters managed by Cloudera Manager run Cloudera Manager Server, monitoring roles, and other Cloudera Server processes as cloudera-scm. Requires keytab file named cmf.keytabbecause name is hard-coded in Cloudera Manager.
Apache Accumulo	accumulo	accumulo	Accumulo processes run as this user.
Apache Flume	flume	flume	The sink that writes to HDFS as user must have write privileges.
Apache HBase	hbase	hbase	The Master and the RegionServer processes run as this user.
HDFS	hdfs	hdfs, hadoop	The NameNode and DataNodes run as this user, and the HDFS root directory as well as the directories used for edit logs should be owned by it.
Apache Hive	hive	hive	The HiveServer2 process and the Hive Metastore processes run as this user. A user must be defined for Hive access to its Metastore DB (for example, MySQL or Postgres) but it can be any identifier and does not correspond to a Unix uid. This is javax.jdo.option.ConnectionUserName in hive-site.xml.
Apache HCatalog	hive	hive	The WebHCat service (for REST access to Hive functionality) runs as the hive user.
HttpFS	httpfs	httpfs	The HttpFS service runs as this user. See HttpFS Security Configuration for instructions on how to generate the merged httpfs-http.keytab file.
Hue	hue	hue	Hue services run as this user.
Hue Load Balancer	apache	apache	The Hue Load balancer has a dependency on the apache2 package that uses the apacheuser name. Cloudera Manager does not run processes using this user ID.
Impala	impala	impala, hive	Impala services run as this user.
Apache Kafka	kafka	kafka	Kafka services run as this user.
Java KeyStore KMS	kms	kms	The Java KeyStore KMS service runs as this user.
Key Trustee KMS	kms	kms	The Key Trustee KMS service runs as this user.
Key Trustee Server	keytrustee	keytrustee	The Key Trustee Server service runs as this user.
Kudu	kudu	kudu	Kudu services run as this user.
MapReduce	mapred	mapred, hadoop	Without Kerberos, the JobTracker and tasks run as this user. The LinuxTaskController binary is owned by this user for Kerberos.
Apache Oozie	oozie	oozie	The Oozie service runs as this user.
Parquet	~	~	No special users.
Apache Pig	~	~	No special users.
Cloudera Search	solr	solr	The Solr processes run as this user.
Apache Spark	spark	spark	The Spark History Server process runs as this user.
Apache Sentry	sentry	sentry	The Sentry service runs as this user.
Apache Sqoop	sqoop	sqoop	This user is only for the Sqoop1 Metastore, a configuration option that is not recommended.
YARN	yarn	yarn, hadoop	Without Kerberos, all YARN services and applications run as this user. The LinuxContainerExecutor binary is owned by this user for Kerberos.

3. 裸机安装指南

(Cloudera Enterprise Reference Architecture for Bare Metal Deployments (PDF))[http://www.cloudera.com/documentation/other/reference-architecture/PDF/cloudera_ref_arch_metal.pdf]

1) java

CM和CDH使用oracle jdk，不支持openjdk。

2) 适当大小的服务器配置

cloudera推荐在生产环境中部署三到四种机器类型：

• Master Node, 运行Hadoop master daemon: NameNode, Standby NameNode, YARN Resource Manager and History Server, the HBase Master daemon, Sentry server, and the Impala StateStore Server and Catalog Server。 同时也是Zookeeper和JournalNodes的安装位置。kudu master servers也会部署在Master Node上。
• Worker Node, Runs the HDFS DataNode, YARN NodeManager, HBase RegionServer, Impala impalad, Search worker daemons and Kudu Tablet Servers.
• Utility Node, Runs Cloudera Manager and the Cloudera Management Services. 也可以部署一个MySQL（或其他）数据库，用于Cloudera Manager, Hive, Sentry 和其他 Hadoop-related projects.
• Edge Node, Contains all client-facing configurations and services, including gateway configurations for HDFS, YARN, Impala, Hive, and HBase. The edge node is also a good place for Hue, Oozie, HiveServer2, and Impala HAProxy. HiveServer2 and Impala HAProxy serve as a gateway to external applications such as Business Intelligence (BI) tools.

4. 扩展阅读

before you install