Postgresql-XL安装部署

机器准备:

3台虚拟机信息:
gtm:a. 版本信息:Linux localhost.localdomain 2.6.32-358.el6.x86_64 #1 SMP Fri Feb 22 00:31:26 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux
b. IP:192.168.100.199

node1:a. 版本信息:Linux localhost.localdomain 2.6.18-194.el5 #1 SMP Fri Apr 2 14:58:14 EDT 2010 x86_64 x86_64 x86_64 GNU/Linux
b. IP:192.168.100.201

node2:a. 版本信息:Linux localhost.localdomain 2.6.18-194.el5 #1 SMP Fri Apr 2 14:58:35 EDT 2010 i686 i686 i386 GNU/Linux
b. IP:192.168.100.202

安装(三台虚机同时按下面步骤安装)

1、下载源码:官方下载网盘下载
2、解压源码

tar xvf postgres-xl-v9.2-src.tar.gz
3、编译安装
./configure --prefix=/opt/pgxl
make
make install
注:编译过程中出现的问题(centos5会出现此问题,6则不会):
'/usr/bin/perl' /bin/collateindex.pl -f -g -i 'bookindex' -o bookindex.sgml HTML.index
Can't open perl script "/bin/collateindex.pl": No such file or directory
make[4]: *** [bookindex.sgml] Error 2
make[4]: Leaving directory `/root/postgres-xl/doc-xc/src/sgml'
make[3]: *** [sql_help.h] Error 2
make[3]: Leaving directory `/root/postgres-xl/src/bin/psql'
make[2]: *** [all-psql-recurse] Error 2
make[2]: Leaving directory `/root/postgres-xl/src/bin'
make[1]: *** [all-bin-recurse] Error 2
make[1]: Leaving directory `/root/postgres-xl/src'
make: *** [all-src-recurse] Error 2
解决方法:
#yum install docbook-style-dsssl
#find / -name collateindex.pl
/usr/bin/collateindex.pl
#cp /usr/bin/collateindex.pl /bin/
节点部署:

新建用户(三台虚机依次操作,注意:一定要用相同的用户,否则会发生用户找不到的问题):

[root@localhost opt]# useradd pgxl
[root@localhost opt]# chown -R pgxl pgxl/
[root@localhost opt]# su pgxl
[pgxl@localhost pgxl]$ ls
bin  data_gtm  include  lib  share
[pgxl@localhost pgxl]$ cd bin/
[pgxl@localhost bin]$ ls
clusterdb   createuser  dropuser  gtm_ctl    initgtm        pg_config       pg_dump         pg_resetxlog  postmaster  vacuumdb
createdb    dropdb      ecpg      gtm_proxy  makesgml       pg_controldata  pg_dumpall      pg_restore    psql
createlang  droplang    gtm       initdb     pg_basebackup  pg_ctl          pg_receivexlog  postgres      reindexdb
[pgxl@localhost bin]$

初始化:

gtm初始化:
[pgxl@localhost bin]$ ./initgtm -D /opt/pgxl/data_gtm/ -Z gtm
The files belonging to this GTM system will be owned by user "pgxl".
This user must also own the server process.


creating directory /opt/pgxl/data_gtm ... ok
creating configuration files ... ok

Success. You can now start the GTM server using:

    ./gtm -D /opt/pgxl/data_gtm
or
    ./gtm_ctl -Z gtm -D /opt/pgxl/data_gtm -l logfile start

[pgxl@localhost bin]$
node1初始化(initdb初始化过程和pg初始化过程一致,这里就不多赘述了):
[pgxl@localhost bin]$ ./initdb -D /opt/pgxl/data/coordinator1/ --nodename coordinator1
[pgxl@localhost bin]$ ./initdb -D /opt/pgxl/data/datanode1/ --nodename datanode1
[pgxl@localhost bin]$ ./initgtm -D /opt/pgxl/data/gtm_proxy1/ -Z gtm_proxy
The files belonging to this GTM system will be owned by user "pgxl".
This user must also own the server process.

creating directory /opt/pgxl/data/gtm_proxy1 ... ok
creating configuration files ... ok

Success. You can now start the GTM proxy server using:

  ./gtm_proxy -D /opt/pgxl/data/gtm_proxy1
or
  ./gtm_ctl -Z gtm_proxy -D /opt/pgxl/data/gtm_proxy1 -l logfile start
node2初始化:
[pgxl@localhost bin]$ ./initdb -D /opt/pgxl/data/coordinator2/ --nodename coordinator2
[pgxl@localhost bin]$ ./initdb -D /opt/pgxl/data/datanode2/ --nodename datanode2
[pgxl@localhost bin]$ ./initgtm -D /opt/pgxl/data/gtm_proxy2/ -Z gtm_proxy
The files belonging to this GTM system will be owned by user "pgxl".
This user must also own the server process.

creating directory /opt/pgxl/data/gtm_proxy2 ... ok
creating configuration files ... ok

Success. You can now start the GTM proxy server using:

  ./gtm_proxy -D /opt/pgxl/data/gtm_proxy2
or
  ./gtm_ctl -Z gtm_proxy -D /opt/pgxl/data/gtm_proxy2 -l logfile start

配置信息:

gtm节点配置(修改项)

[pgxl@localhost data_gtm]$ vi gtm.conf
nodename = 'gtm'                    
listen_addresses = '*'
keepalives_idle = 60		
keepalives_interval = 10	
keepalives_count = 10
node1节点配置(修改项):
[pgxl@localhost gtm_proxy1]$ vi gtm_proxy.conf
nodename = 'gtm_proxy1'                                 
                                                               
listen_addresses = '*'                 
port = 6661                                    
gtm_host = '192.168.100.199'                                 
gtm_port = 6666
keepalives_idle = 60
keepalives_interval = 10
keepalives_count = 10
[pgxl@localhost coordinator1]$ vi postgresql.conf 
listen_addresses = '*'
#gtm_host = 'localhost'                
gtm_port = 6661               
pgxc_node_name = 'coordinator1'  pooler_port = 6667
[pgxl@localhost coordinator1]$ vi pg_hba.conf
host    all             all             192.168.100.0/24            trust
[pgxl@localhost datanode1]$ vi postgresql.conf
listen_addresses = '*'
port = 15432
#gtm_host = 'localhost' 
gtm_port = 6661             
pgxc_node_name = 'datanode1'
pooler_port = 6668
[pgxl@localhost datanode1]$ vi pg_hba.conf
host    all             all             192.168.100.0/24            trust
node2节点配置(修改项):
[pgxl@localhost gtm_proxy2]$ vi gtm_proxy.conf
nodename = 'gtm_proxy2'                                 
                                                               
listen_addresses = '*'                 
port = 6662                                    
gtm_host = '192.168.100.199'                                 
gtm_port = 6666
keepalives_idle = 60
keepalives_interval = 10
keepalives_count = 10
[pgxl@localhost coordinator2]$ vi postgresql.conf 
listen_addresses = '*'
#gtm_host = 'localhost'                
gtm_port = 6662               
pgxc_node_name = 'coordinator2'  
pooler_port = 6667
[pgxl@localhost coordinator2]$ vi pg_hba.conf
host    all             all             192.168.100.0/24            trust
[pgxl@localhost datanode2]$ vi postgresql.conf
listen_addresses = '*'
port = 15432
#gtm_host = 'localhost' 
gtm_port = 6662             
pgxc_node_name = 'datanode2'
pooler_port = 6668
[pgxl@localhost datanode2]$ vi pg_hba.conf
host    all             all             192.168.100.0/24            trust

服务启动:

gtm节点:

[pgxl@localhost bin]$ export PGHOME=/opt/pgxl/
[pgxl@localhost bin]$ export PATH=$PGHOME/bin:$PATH
[pgxl@localhost bin]$ ./gtm_ctl start -Z gtm -D /opt/pgxl/data_gtm/
server starting
[pgxl@localhost bin]$ tail ../data_gtm/gtm.log 
1:139738148718336:2015-06-09 14:23:08.247 CST -LOG:  Starting GTM server at (*:6666) -- control file /opt/pgxl/data_gtm/gtm.control
LOCATION:  main,main.c:601
1:139738148718336:2015-06-09 14:23:08.247 CST -LOG:  Restoring last GXID to 10000
	
LOCATION:  GTM_RestoreTxnInfo,gtm_txn.c:2673
1:139738148718336:2015-06-09 14:23:08.247 CST -LOG:  Started to run as GTM-Active.
LOCATION:  main,main.c:682
node1节点:
[pgxl@localhost bin]$ export PGHOME=/opt/pgxl/
[pgxl@localhost bin]$ export PATH=$PGHOME/bin:$PATH
[pgxl@localhost bin]$ ./gtm_ctl start -Z gtm_proxy -D ../data/gtm_proxy1/
server starting
[pgxl@localhost bin]$ tail ../data/gtm_proxy1/gtm_proxy.log 
1:47078985160944:2015-06-08 23:27:22.567 PDT -LOG:  Starting GTM proxy at (*:6661)
LOCATION:  main,proxy_main.c:805
[pgxl@localhost bin]$ ./pg_ctl start -Z datanode -D ../data/datanode1/
[pgxl@localhost bin]$ LOG:  database system was shut down at 2015-06-08 22:35:18 PDT
LOG:  autovacuum launcher started
LOG:  database system is ready to accept connections

[pgxl@localhost bin]$ ./pg_ctl start -Z coordinator -D ../data/coordinator1/
[pgxl@localhost bin]$ LOG:  database system was shut down at 2015-06-08 22:34:46 PDT
LOG:  autovacuum launcher started
LOG:  database system is ready to accept connections
node2节点:
[pgxl@localhost bin]$ export PGHOME=/opt/pgxl/
[pgxl@localhost bin]$ export PATH=$PGHOME/bin:$PATH
[pgxl@localhost bin]$ ./gtm_ctl start -Z gtm_proxy -D ../data/gtm_proxy2/
server starting
[pgxl@localhost bin]$ tail ../data/gtm_proxy2/gtm_proxy.log 
1:47078985160944:2015-06-08 23:27:22.567 PDT -LOG:  Starting GTM proxy at (*:6662)
LOCATION:  main,proxy_main.c:805
[pgxl@localhost bin]$ ./pg_ctl start -Z datanode -D ../data/datanode2/
[pgxl@localhost bin]$ LOG:  database system was shut down at 2015-06-08 22:35:18 PDT
LOG:  autovacuum launcher started
LOG:  database system is ready to accept connections

[pgxl@localhost bin]$ ./pg_ctl start -Z coordinator -D ../data/coordinator2/
[pgxl@localhost bin]$ LOG:  database system was shut down at 2015-06-08 22:34:46 PDT
LOG:  autovacuum launcher started
LOG:  database system is ready to accept connections
注册:

node1:

协调节点:

[pgxl@localhost bin]$ ./psql -p 5432 -d postgres
psql (PGXL 9.2.0,based on PG 9.2.4 (Postgres-XL 9.2.0))
Type "help" for help.

postgres=# CREATE NODE coordinator2 WITH (TYPE='coordinator',HOST='192.168.100.202',PORT=5432);
CREATE NODE
postgres=# CREATE NODE datanode1 WITH (TYPE='datanode',HOST='192.168.100.201',PORT=15432);
CREATE NODE
postgres=# CREATE NODE datanode2 WITH (TYPE='datanode',PORT=15432);
CREATE NODE

postgres=# select * from pgxc_node;
 node_name  | node_type | node_port |  node_host  | nodeis_primary | nodeis_preferred |  node_id 
--------------+-----------+-----------+-----------------+----------------+------------------+-------------
coordinator1 | C     |   5432 | localhost    | f       | f        | 1938253334
coordinator2 | C     |   5432 | 192.168.100.202 | f       | f        | -2089598990
datanode1  | D     |   15432 | 192.168.100.201 | f       | f        |  888802358
datanode2  | D     |   15432 | 192.168.100.202 | f       | f        | -905831925

select pgxc_pool_reload();
 pgxc_pool_reload 
------------------
 t
(1 row)

[pgxl@localhost bin]$

数据节点:

[pgxl@localhost bin]$ ./psql -p 15432 -d postgres
psql (PGXL 9.2.0,based on PG 9.2.4 (Postgres-XL 9.2.0))
Type "help" for help.

postgres=#CREATE NODE coordinator1 WITH (TYPE='coordinator',PORT=5432);
CREATE NODE
postgres=#CREATE NODE coordinator2 WITH (TYPE='coordinator',PORT=5432);
CREATE NODE
postgres=#ALTER NODE datanode1 WITH (TYPE='datanode',HOST='localhost',PORT=15432);
ALTER NODE
postgres=#CREATE NODE datanode2 WITH (TYPE='datanode',PORT=15432);
CREATE NODE

postgres=# select * from pgxc_node;
 node_name  | node_type | node_port |  node_host  | nodeis_primary | nodeis_preferred |  node_id 
--------------+-----------+-----------+-----------------+----------------+------------------+-------------
coordinator1 | C     |   5432 | 192.168.100.201 | f       | f        | 1938253334
coordinator2 | C     |   5432 | 192.168.100.202 | f       | f        | -2089598990
datanode1  | D     |   15432 | localhost    | f       | f        |  888802358
datanode2  | D     |   15432 | 192.168.100.202 | f       | f        | -905831925
(4 rows)

select pgxc_pool_reload();
 pgxc_pool_reload 
------------------
 t
(1 row)

[pgxl@localhost bin]$

node2:

协调节点:

[pgxl@localhost bin]$ ./psql -p 5432 -d postgres
psql (PGXL 9.2.0,based on PG 9.2.4 (Postgres-XL 9.2.0))
Type "help" for help.

postgres=# CREATE NODE coordinator1 WITH (TYPE='coordinator',PORT=15432);
CREATE NODE

postgres=# select * from pgxc_node;
 node_name  | node_type | node_port |  node_host  | nodeis_primary | nodeis_preferred |  node_id 
--------------+-----------+-----------+-----------------+----------------+------------------+-------------
coordinator2 | C     |   5432 | localhost    | f       | f        | -2089598990
coordinator1 | C     |   5432 | 192.168.100.201 | f       | f        | 1938253334
datanode1  | D     |   15432 | 192.168.100.201 | f       | f        |  888802358
datanode2  | D     |   15432 | 192.168.100.202 | f       | f        | -905831925

select pgxc_pool_reload();
 pgxc_pool_reload 
------------------
 t
(1 row)

[pgxl@localhost bin]$

数据节点:

[pgxl@localhost bin]$ ./psql -p 15432 -d postgres
psql (PGXL 9.2.0,PORT=5432);
CREATE NODE
postgres=#ALTER NODE datanode2 WITH (TYPE='datanode',PORT=15432);
ALTER NODE
postgres=#CREATE NODE datanode1 WITH (TYPE='datanode',PORT=15432);
CREATE NODE

postgres=# select * from pgxc_node;
 node_name  | node_type | node_port |  node_host  | nodeis_primary | nodeis_preferred |  node_id 
--------------+-----------+-----------+-----------------+----------------+------------------+-------------
coordinator1 | C     |   5432 | 192.168.100.201 | f       | f        | 1938253334
coordinator2 | C     |   5432 | 192.168.100.202 | f       | f        | -2089598990
datanode2  | D     |   15432 | localhost    | f       | f        | -905831925
datanode1  | D     |   15432 | 192.168.100.201 | f       | f        | 888802358
(4 rows)

select pgxc_pool_reload();
 pgxc_pool_reload 
------------------
 t
(1 row)

[pgxl@localhost bin]$

搭建后效果:


node1:
postgres=# create table test(t1 int,t2 text);
LOG:  Will fall back to local snapshot for XID = 16757,source = 0,gxmin = 0,autovac launch = 0,autovac = 0,normProcMode = 0,postEnv = 1
CREATE TABLE
postgres=# insert into test values (1,'a');
INSERT 0 1
postgres=# insert into test values (2,'b');
INSERT 0 1
postgres=# execute direct on (datanode1) $$ select * from test $$;
 t1 | t2 
----+----
  1 | a
  2 | b
(2 rows)

postgres=# select * from test ;
 t1 | t2 
----+----
  1 | a
  2 | b
(2 rows)

postgres=# 

node2:
postgres=# select * from test ;
LOG:  Will fall back to local snapshot for XID = 16789,postEnv = 1
 t1 | t2 
----+----
  1 | a
  2 | b
(2 rows)

postgres=# execute direct on (datanode2) $$ select * from test $$;
 t1 | t2 
----+----
(0 rows)

postgres=# insert into test values (3,'c');
INSERT 0 1
postgres=# execute direct on (datanode2) $$ select * from test $$;
 t1 | t2 
----+----
  3 | c
(1 row)

postgres=# select * from test ;
 t1 | t2 
----+----
  1 | a
  2 | b
  3 | c
(3 rows)

postgres=#

node1:
postgres=# select * from test ;
 t1 | t2 
----+----
  1 | a
  2 | b
  3 | c
(3 rows)

postgres=# execute direct on (datanode1) $$ select * from test $$;
 t1 | t2 
----+----
  1 | a
  2 | b
(2 rows)

postgres=# execute direct on (datanode2) $$ select * from test $$;
 t1 | t2 
----+----
  3 | c
(1 row)

postgres=#

发现bug(最新代码没有,官网发布的tar包有bug,通过git下载暂时没有问题)

postgres=# alter node datanode2 with(TYPE='datanode',PORT=15432);
The connection to the server was lost. Attempting reset: LOG:  server process (PID 14999) was terminated by signal 11: Segmentation fault
DETAIL:  Failed process was running: alter node datanode2 with(TYPE='datanode',PORT=15432);
LOG:  terminating any other active server processes
WARNING:  terminating connection because of crash of another server process
DETAIL:  The postmaster has commanded this server process to roll back the current transaction and exit,because another server process exited abnormally and possibly corrupted shared memory.
HINT:  In a moment you should be able to reconnect to the database and repeat your command.
FATAL:  the database system is in recovery mode
Failed.
!> LOG:  all server processes terminated; reinitializing
LOG:  database system was interrupted; last known up at 2015-06-10 20:09:24 PDT
LOG:  database system was not properly shut down; automatic recovery in progress
LOG:  record with zero length at 0/18A6C40
LOG:  redo is not required
LOG:  autovacuum launcher started
LOG:  database system is ready to accept connections

!> 
!> 
!> 
!>

版权声明:本文内容由互联网用户自发贡献,该文观点与技术仅代表作者本人。本站仅提供信息存储空间服务,不拥有所有权,不承担相关法律责任。如发现本站有涉嫌侵权/违法违规的内容, 请发送邮件至 dio@foxmail.com 举报,一经查实,本站将立刻删除。

相关推荐


文章浏览阅读601次。Oracle的数据导入导出是一项基本的技能,但是对于懂数据库却不熟悉Oracle的同学可能会有一定的障碍。正好在最近的一个项目中碰到了这样一个任务,于是研究了一下Oracle的数据导入导出,在这里跟大家分享一下。......_oracle 迁移方法 对比
文章浏览阅读553次。开头还是介绍一下群,如果感兴趣polardb ,mongodb ,mysql ,postgresql ,redis 等有问题,有需求都可以加群群内有各大数据库行业大咖,CTO,可以解决你的问题。加群请联系 liuaustin3 ,在新加的朋友会分到2群(共700多人左右 1 + 2)。最近我们在使用MYSQL 8 的情况下(8.025)在数据库运行中出现一个问题 参数prefer_order_i..._mysql prefer_ordering_index
文章浏览阅读3.5k次,点赞3次,收藏7次。折腾了两个小时多才成功连上,在这分享一下我的经验,也仅仅是经验分享,有不足的地方欢迎大家在评论区补充交流。_navicat连接opengauss
文章浏览阅读2.7k次。JSON 代表 JavaScript Object Notation。它是一种开放标准格式,将数据组织成中详述的键/值对和数组。_postgresql json
文章浏览阅读2.9k次,点赞2次,收藏6次。navicat 连接postgresql 注:navicat老版本可能报错。1.在springboot中引入我们需要的依赖以及相应版本。用代码生成器生成代码后,即可进行增删改查(略)安装好postgresql 略。更改配置信息(注释中有)_mybatisplus postgresql
文章浏览阅读1.4k次。postgre进阶sql,包含分组排序、JSON解析、修改、删除、更新、强制踢出数据库所有使用用户、连表更新与删除、获取今年第一天、获取近12个月的年月、锁表处理、系统表使用(查询所有表和字段及注释、查询表占用空间)、指定数据库查找模式search_path、postgre备份及还原_pgsql分组取每组第一条
文章浏览阅读3.3k次。上一篇我们学习了日志清理,日志清理虽然解决了日志膨胀的问题,但就无法再恢复检查点之前的一致性状态。因此,我们还需要日志归档,pg的日志归档原理和Oracle类似,不过归档命令需要自己配置。以下代码在postmaster.c除了开启归档外,还需要保证wal_level不能是MINIMAL状态(因为该状态下有些操作不会记录日志)。在db启动时,会同时检查archive_mode和wal_level。以下代码也在postmaster.c(PostmasterMain函数)。......_postgresql archive_mode
文章浏览阅读3k次。系统:ubuntu22.04.3目的:利用向日葵实现windows远程控制ubuntu。_csdn局域网桌面控制ubuntu
文章浏览阅读1.6k次。表分区是解决一些因单表过大引用的性能问题的方式,比如某张表过大就会造成查询变慢,可能分区是一种解决方案。一般建议当单表大小超过内存就可以考虑表分区了。1,继承式分区,分为触发器(trigger)和规则(rule)两种方式触发器的方式1)创建表CREATE TABLE "public"."track_info_trigger_partition" ( "id" serial, "object_type" int2 NOT NULL DEFAULT 0, "object_name..._pg数据表分区的实现
文章浏览阅读3.3k次。物联网平台开源的有几个,就我晓得的有、、thingskit、JetLink、DG-iot(还有其他开源的,欢迎在评论区留言哦!),然后重点分析了下ThingsBoard、ThingsPanel和JetLink,ThingsBoard和Jetlinks是工程师思维产品,可以更多的通过配置去实现开发的目的,ThingsPanel是业务人员思路产品,或者开发或者用,避免了复杂的配置带来的较高学习门槛。ThingsBoard和Jetlinks是Java技术体系的,ThingsPanel是PHP开发的。_jetlinks和thingsboard
文章浏览阅读3.8k次。PostgreSQL 数据类型转换_pgsql数字转字符串
文章浏览阅读7k次,点赞3次,收藏14次。在做数据统计页面时,总会遇到统计某段时间内,每天、每月、每年的数据视图(柱状图、折线图等)。这些统计数据一眼看过去也简单呀,不就是按照时间周期(天、月、年)对统计数据进行分个组就完了嘛?但是会有一个问题,简单的写个sql对周期分组,获取到的统计数据是缺失的,即没有数据的那天,整条记录也都没有了。如下图需求:以当前月份(2023年2月)为起点,往后倒推一年,查询之前一年里每个月的统计数据。可见图中的数据其实是缺少的,这条sql只查询到了有数据的月份(23年的1月、2月,22年的12月)_如何用一条sql查出按年按月按天的汇总
文章浏览阅读3.8k次,点赞66次,收藏51次。PostgreSQL全球开发小组与2022年10月13日,宣布发布PostgreSQL15,这是世界上最先进的开源数据库的最新版本_mysql8 postgresql15
文章浏览阅读1.3k次。上文介绍了磁盘管理器中VFD的实现原理,本篇将从上层角度讲解磁盘管理器的工作细节。_smgrrelationdata
文章浏览阅读1.1k次。PostgreSQL设置中文语言界面和局域网访问_postgressql汉化
文章浏览阅读4.2k次。PostgreSQL 修改数据存储路径_如何设置postgresql 数据目录
文章浏览阅读4.7k次。在项目中用到了多数据源,在连接postgres数据库时,项目启动报错,说数据库连接错误,说dual不存在,网上好多教程都是说数据库查询的时候的大小写问题,而这个仅仅是连接,咋鞥却处理方法是修改application-dev.yml中的配置文件.项目中的druid参数是这样的:确实在配置文件中有个查询语句。_relation "dual" does not exist
文章浏览阅读4.9k次。PostgreSQL是一款强大的关系型数据库,但在实际使用过程中,许多用户经常会遇到慢SQL的问题。这些问题不仅会降低数据库性能,还会直接影响业务流程和用户体验。因此,本文将会深入分析PostgreSQL慢SQL的原因和优化方案,帮助用户更好地利用这个优秀的数据库系统。无论你是初学者还是专业开发者,本文都将为你提供实用的技巧和方法,让你的PostgreSQL数据库始终保持高效快速。_postgresql数据库优化
文章浏览阅读1.6k次。Linux配置postgresql开机自启_linux 启动pgsql
文章浏览阅读2k次。本篇介绍如何在centos7系统搭建一个postgresql主备集群实现最近的HA(高可用)架构。后续更高级的HA模式都是基于这个最基本的主备搭建。_postgresql主备