GreenPlum的那些事《二》——GreenPlun启动报错

在配置好GreenPlum过程中，需要注意一下几点：

  1. 找好安装包
  2. 安装依赖，Gun版本需要注意
  3. 详细安装步骤参考Greenplum安装介绍

自己是电脑虚拟机安装的中间出现很多报错，度娘加经验解决。最后启动还是报错，发现是资源不足，所以关闭了mirror，减少了segment。然后启动还是报错：

>     > 20190122:20:23:14:003437 gpstart:mpp01:gpadmin-[INFO]:-Commencing parallel segment instance
> startup, please wait...
>     ... 
>     20190122:20:23:17:003437 gpstart:mpp01:gpadmin-[INFO]:-Process results...
>     20190122:20:23:17:003437 gpstart:mpp01:gpadmin-[INFO]:-----------------------------------------------------
>     20190122:20:23:17:003437 gpstart:mpp01:gpadmin-[INFO]:-   Successful segment starts                                            =
> 3
>     20190122:20:23:17:003437 gpstart:mpp01:gpadmin-[INFO]:-   Failed segment starts                                                = 0
>     20190122:20:23:17:003437 gpstart:mpp01:gpadmin-[INFO]:-   Skipped segment starts (segments are marked down in configuration)   = 0
>     20190122:20:23:17:003437 gpstart:mpp01:gpadmin-[INFO]:-----------------------------------------------------
>     20190122:20:23:17:003437 gpstart:mpp01:gpadmin-[INFO]:-Successfully started 3 of 3 segment
> instances 
>     20190122:20:23:17:003437 gpstart:mpp01:gpadmin-[INFO]:-----------------------------------------------------
>     20190122:20:23:17:003437 gpstart:mpp01:gpadmin-[INFO]:-Starting Master instance mpp01 directory /opt/greenplum/data/master/gpseg-1 
>     20190122:20:23:18:003437 gpstart:mpp01:gpadmin-[INFO]:-Command pg_ctl reports Master mpp01 instance active
>     20190122:20:23:19:003437 gpstart:mpp01:gpadmin-[WARNING]:-FATAL:  DTM initialization: failure during startup recovery, retry failed,
> check segment status (cdbtm.c:1517)
>     20190122:20:23:19:003437 gpstart:mpp01:gpadmin-[INFO]:-No standby master configured.  skipping...
>     20190122:20:23:19:003437 gpstart:mpp01:gpadmin-[INFO]:-Check status of database with gpstate utility

查看日志没有找到什么原因，后台度娘发现了，原来是防火墙没有关闭。那么就来关闭防火墙：

/etc/init.d/iptables stop
[root@mpp03 ~]# /etc/init.d/iptables stop  
iptables: Setting chains to policy ACCEPT: filter          [  OK  ]
iptables: Flushing firewall rules:                         [  OK  ]
iptables: Unloading modules:                               [  OK  ]

在3台机器上都关闭防火墙操作，然后启动

gpstart

    20190122:20:28:01:004200 gpstart:mpp01:gpadmin-[INFO]:-Process results...
20190122:20:28:01:004200 gpstart:mpp01:gpadmin-[INFO]:-----------------------------------------------------
20190122:20:28:01:004200 gpstart:mpp01:gpadmin-[INFO]:-   Successful segment starts                                            = 3
20190122:20:28:01:004200 gpstart:mpp01:gpadmin-[INFO]:-   Failed segment starts                                                = 0
20190122:20:28:01:004200 gpstart:mpp01:gpadmin-[INFO]:-   Skipped segment starts (segments are marked down in configuration)   = 0
20190122:20:28:01:004200 gpstart:mpp01:gpadmin-[INFO]:-----------------------------------------------------
20190122:20:28:01:004200 gpstart:mpp01:gpadmin-[INFO]:-Successfully started 3 of 3 segment instances 
20190122:20:28:01:004200 gpstart:mpp01:gpadmin-[INFO]:-----------------------------------------------------
20190122:20:28:01:004200 gpstart:mpp01:gpadmin-[INFO]:-Starting Master instance mpp01 directory /opt/greenplum/data/master/gpseg-1 
20190122:20:28:02:004200 gpstart:mpp01:gpadmin-[INFO]:-Command pg_ctl reports Master mpp01 instance active
20190122:20:28:02:004200 gpstart:mpp01:gpadmin-[INFO]:-No standby master configured.  skipping...
20190122:20:28:02:004200 gpstart:mpp01:gpadmin-[INFO]:-Database successfully started

启动成功！

psql
[gpadmin@mpp01 ~]$ psql
psql (8.3.23)
Type "help" for help.

gp_sydb=#

启动成功了，开始下面的学习吧