因缺失增量信息而导致recoverseg恢复失败的情况

最新推荐文章于 2023-12-09 19:50:01 发布
最新推荐文章于 2023-12-09 19:50:01 发布
阅读量4.1k 收藏 1
点赞数
分类专栏： 数据库 greenplum 文章标签： greenplum 数据库 gprecoverseg 系统恢复
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
本文链接：https://blog.csdn.net/aabc012/article/details/46336681
版权

以下只记录重要信息。第1，2，3部分是描述错误产生的原因。第4 部分是解决方法。

1，对gpseg1的primary 进行更名操作让其发生主备切换。

-- 对mirror进行更名操作并不会让其马上失联，gp_segment_connect_timeout中设定的默认值为10分钟，更改回去后还会接上。

[gpadmin@sdw2primary]$ mv gpseg1 gpseg11

[gpadmin@sdw2primary]$ mv gpseg11 gpseg1

 

--gpseg1的主备切换了,mirror记录变化。

[gpadmin@mdw ~]$ gpstate -s

20150602:21:59:05:053347gpstate:mdw:gpadmin-[INFO]:-----------------------------------------------------

20150602:21:59:05:053347gpstate:mdw:gpadmin-[INFO]:-   SegmentInfo

20150602:21:59:05:053347gpstate:mdw:gpadmin-[INFO]:-     Hostname                          = sdw1

20150602:21:59:05:053347gpstate:mdw:gpadmin-[INFO]:-     Address                          = sdw1

20150602:21:59:05:053347gpstate:mdw:gpadmin-[INFO]:-     Datadir                          = /data1/mirror/gpseg1

20150602:21:59:05:053347gpstate:mdw:gpadmin-[INFO]:-     Port                             = 50000

20150602:21:59:05:053347gpstate:mdw:gpadmin-[INFO]:-   MirroringInfo

20150602:21:59:05:053347gpstate:mdw:gpadmin-[INFO]:-      Currentrole                      = Primary

20150602:21:59:05:053347gpstate:mdw:gpadmin-[INFO]:-     Preferred role                   = Mirror

20150602:21:59:05:053347gpstate:mdw:gpadmin-[INFO]:-      Mirrorstatus                     = ChangeTracking

20150602:21:59:05:053347gpstate:mdw:gpadmin-[INFO]:-   ChangeTracking Info

20150602:21:59:05:053347gpstate:mdw:gpadmin-[INFO]:-      Changetracking data size         = 100 MB

20150602:21:59:05:053347gpstate:mdw:gpadmin-[INFO]:-   Status

20150602:21:59:05:053347gpstate:mdw:gpadmin-[INFO]:-     PID                              = 10792

20150602:21:59:05:053347gpstate:mdw:gpadmin-[INFO]:-     Configuration reports status as  = Up

20150602:21:59:05:053347gpstate:mdw:gpadmin-[INFO]:-     Database status                  = Up

20150602:21:59:05:053347gpstate:mdw:gpadmin-[INFO]:-----------------------------------------------------

20150602:21:59:05:053347gpstate:mdw:gpadmin-[INFO]:-   SegmentInfo

20150602:21:59:05:053347gpstate:mdw:gpadmin-[INFO]:-     Hostname                          = sdw2

20150602:21:59:05:053347gpstate:mdw:gpadmin-[INFO]:-     Address                          = sdw2

20150602:21:59:05:053347gpstate:mdw:gpadmin-[INFO]:-     Datadir                          = /data1/primary/gpseg1

20150602:21:59:05:053347gpstate:mdw:gpadmin-[INFO]:-     Port                             = 40000

20150602:21:59:05:053347gpstate:mdw:gpadmin-[INFO]:-   MirroringInfo

20150602:21:59:05:053347gpstate:mdw:gpadmin-[INFO]:-      Currentrole                      = Mirror

20150602:21:59:05:053347gpstate:mdw:gpadmin-[INFO]:-     Preferred role                   = Primary

20150602:21:59:05:053347gpstate:mdw:gpadmin-[WARNING]:-   Mirrorstatus                     = Out ofSync             <<<<<<<<

20150602:21:59:05:053347gpstate:mdw:gpadmin-[INFO]:-   Status

20150602:21:59:05:053347gpstate:mdw:gpadmin-[WARNING]:-  PID                              = Not found              <<<<<<<<

20150602:21:59:05:053347gpstate:mdw:gpadmin-[WARNING]:-  Configuration reports status as  = Down                   <<<<<<<<

20150602:21:59:05:053347gpstate:mdw:gpadmin-[WARNING]:-   Segmentstatus                    = Down inconfiguration   <<<<<<<<

20150602:21:59:05:053347gpstate:mdw:gpadmin-[WARNING]:-*****************************************************

20150602:21:59:05:053347gpstate:mdw:gpadmin-[WARNING]:-Warnings have been generated during statusprocessing

20150602:21:59:05:053347gpstate:mdw:gpadmin-[WARNING]:-Check log file or review screen output

20150602:21:59:05:053347gpstate:mdw:gpadmin-[WARNING]:-*****************************************************

 

2，--但是对数据库进行了几次操作后，我并不想要gpseg1中mirror的数据，我还是想让mirror和原先的primary同步。（嗯，就是这么任性！）

--进入master模式修改系统表，让mirror和primary的角色和状态对调一下。

[gpadmin@mdw ~]$ gpstop-a

[gpadmin@mdw ~]$ gpstart -m

[gpadmin@mdw ~]$ PGOPTIONS="-cgp_session_role=utility" psql

psql (8.2.15)

Type"help" for help.

 

testDB=# set allow_system_table_mods='dml';

SET

testDB=# select * from gp_segment_configuration;

 dbid | content | role | preferred_role | mode| status | port  | hostname | address |replication_port | san_mounts

------+---------+------+----------------+------+--------+-------+----------+---------+------------------+------------

    2 |      0 | p    | p              | s    | u     | 40000 | sdw1     | sdw1    |           41000 |

    4 |      0 | m    | m              | s    | u     | 50000 | sdw2     | sdw2    |           51000 |

    3 |      1 | m    | p              | s    | d     | 40000 | sdw2     | sdw2    |           41000 |

    5 |      1 | p    | m              | c    | u     | 50000 | sdw1     | sdw1    |           51000 |

    1 |     -1 | p    | p              | s    | u     |  5432 | mdw      | mdw    |                  |

(5 rows)

--

testDB=# update gp_segment_configuration setrole='p',mode='c',status='u' where dbid=3;

UPDATE 1

testDB=# update gp_segment_configuration setrole='m',mode='s',status='d' where dbid=5;

UPDATE 1

testDB=# select * from gp_segment_configuration;

 dbid | content | role | preferred_role | mode| status | port  | hostname | address |replication_port | san_mounts

------+---------+------+----------------+------+--------+-------+----------+---------+------------------+------------

    2 |      0 | p    | p              | s    | u     | 40000 | sdw1     | sdw1    |           41000 |

    4 |      0 | m    | m              | s    | u     | 50000 | sdw2     | sdw2    |           51000 |

    3 |      1 | p    | p              | c    | u     | 40000 | sdw2     | sdw2    |           41000 |

    5 |      1 | m    | m              | s    | d     | 50000 | sdw1     | sdw1    |           51000 |

    1 |     -1 | p    | p              | s    | u     |  5432 | mdw      | mdw    |                  |

(5 rows)

 

-- 关掉master模式。

[gpadmin@mdw ~]$ gpstop -M fast

-- 启动数据库，有个节点起不来。

[gpadmin@mdw ~]$ gpstart -a

20150602:22:08:00:055767gpstart:mdw:gpadmin-[INFO]:-Process results...

20150602:22:08:00:055767gpstart:mdw:gpadmin-[INFO]:-----------------------------------------------------

20150602:22:08:00:055767gpstart:mdw:gpadmin-[INFO]:-   Successfulsegment starts                                            = 3

20150602:22:08:00:055767gpstart:mdw:gpadmin-[INFO]:-   Failedsegment starts                                               = 0

20150602:22:08:00:055767gpstart:mdw:gpadmin-[WARNING]:-Skipped segment starts (segments are marked downin configuration)   = 1   <<<<<<<<

 

3，--恢复一下，但肯定失败，因为primary在启动之前根本就没记录到任何变化，而那时mirror相对还有较多的记录。

[gpadmin@mdw ~]$ gprecoverseg

20150602:22:16:40:062823gprecoverseg:mdw:gpadmin-[INFO]:-Starting gprecoverseg with args:

20150602:22:16:40:062823gprecoverseg:mdw:gpadmin-[INFO]:-local Greenplum Version: 'postgres (GreenplumDatabase) 4.3.5.1 build 1'

20150602:22:16:40:062823gprecoverseg:mdw:gpadmin-[INFO]:-master Greenplum Version: 'PostgreSQL 8.2.15(Greenplum Database 4.3.5.1 build 1) on x86_64-unknown-linux-gnu, compiled byGCC gcc (GCC) 4.4.2 compiled on May 14 2015 14:07:14'

20150602:22:16:40:062823gprecoverseg:mdw:gpadmin-[INFO]:-Checking if segments are ready

20150602:22:16:40:062823gprecoverseg:mdw:gpadmin-[INFO]:-Obtaining Segment details from master...

20150602:22:16:40:062823gprecoverseg:mdw:gpadmin-[INFO]:-Obtaining Segment details from master...

20150602:22:16:40:062823gprecoverseg:mdw:gpadmin-[INFO]:-Greenplum instance recovery parameters

20150602:22:16:40:062823gprecoverseg:mdw:gpadmin-[INFO]:----------------------------------------------------------

20150602:22:16:40:062823gprecoverseg:mdw:gpadmin-[INFO]:-Recovery type              = Standard

20150602:22:16:40:062823gprecoverseg:mdw:gpadmin-[INFO]:----------------------------------------------------------

20150602:22:16:40:062823gprecoverseg:mdw:gpadmin-[INFO]:-Recovery 1 of 1

20150602:22:16:40:062823gprecoverseg:mdw:gpadmin-[INFO]:----------------------------------------------------------

20150602:22:16:40:062823gprecoverseg:mdw:gpadmin-[INFO]:-  Synchronization mode                        = Incremental

20150602:22:16:40:062823gprecoverseg:mdw:gpadmin-[INFO]:-  Failed instance host                        = sdw1

20150602:22:16:40:062823gprecoverseg:mdw:gpadmin-[INFO]:-   Failedinstance address                     =sdw1

20150602:22:16:40:062823gprecoverseg:mdw:gpadmin-[INFO]:-  Failed instance directory                   = /data1/mirror/gpseg1

20150602:22:16:40:062823gprecoverseg:mdw:gpadmin-[INFO]:-  Failed instance port                        = 50000

20150602:22:16:40:062823gprecoverseg:mdw:gpadmin-[INFO]:-  Failed instance replication port            = 51000

20150602:22:16:40:062823gprecoverseg:mdw:gpadmin-[INFO]:-  Recovery Source instance host               = sdw2

20150602:22:16:40:062823gprecoverseg:mdw:gpadmin-[INFO]:-  Recovery Source instance address            = sdw2

20150602:22:16:40:062823gprecoverseg:mdw:gpadmin-[INFO]:-  Recovery Source instance directory          = /data1/primary/gpseg1

20150602:22:16:40:062823gprecoverseg:mdw:gpadmin-[INFO]:-  Recovery Source instance port               = 40000

20150602:22:16:40:062823gprecoverseg:mdw:gpadmin-[INFO]:-  Recovery Source instance replication port   = 41000

20150602:22:16:40:062823gprecoverseg:mdw:gpadmin-[INFO]:-  Recovery Target                             = in-place

20150602:22:16:40:062823gprecoverseg:mdw:gpadmin-[INFO]:----------------------------------------------------------

 

Continue withsegment recovery procedure Yy|Nn (default=N):

> y

20150602:22:16:46:062823gprecoverseg:mdw:gpadmin-[INFO]:-1 segment(s) to recover

20150602:22:16:46:062823gprecoverseg:mdw:gpadmin-[INFO]:-Ensuring 1 failed segment(s) are stopped

 

20150602:22:16:47:062823gprecoverseg:mdw:gpadmin-[INFO]:-Ensuring that shared memory is cleaned up forstopped segments

updating flat files

20150602:22:16:52:062823gprecoverseg:mdw:gpadmin-[INFO]:-Updating configuration with new mirrors

20150602:22:16:52:062823gprecoverseg:mdw:gpadmin-[INFO]:-Updating mirrors

.

20150602:22:16:53:062823gprecoverseg:mdw:gpadmin-[INFO]:-Starting mirrors

20150602:22:16:53:062823gprecoverseg:mdw:gpadmin-[INFO]:-Commencing parallel primary and mirror segmentinstance startup, please wait...

..

20150602:22:16:55:062823gprecoverseg:mdw:gpadmin-[INFO]:-Process results...

20150602:22:16:55:062823gprecoverseg:mdw:gpadmin-[INFO]:-Updating configuration to mark mirrors up

20150602:22:16:55:062823gprecoverseg:mdw:gpadmin-[INFO]:-Updating primaries

20150602:22:16:55:062823gprecoverseg:mdw:gpadmin-[INFO]:-Commencing parallel primary conversion of 1segments, please wait...

..

20150602:22:16:57:062823gprecoverseg:mdw:gpadmin-[INFO]:-Process results...

20150602:22:16:57:062823gprecoverseg:mdw:gpadmin-[WARNING]:-Failed to inform primary segment of updatedmirroring state.  Segment:sdw2:/data1/primary/gpseg1:content=1:dbid=3:mode=r:status=u: REASON: Conversionfailed.  stdout:""  stderr:"failure: Error: MirroringFailurefailure: Error: MirroringFailure "

20150602:22:16:57:062823gprecoverseg:mdw:gpadmin-[INFO]:-Done updating primaries

20150602:22:16:57:062823gprecoverseg:mdw:gpadmin-[INFO]:-******************************************************************

20150602:22:16:57:062823gprecoverseg:mdw:gpadmin-[INFO]:-Updating segments for resynchronization iscompleted.

20150602:22:16:57:062823gprecoverseg:mdw:gpadmin-[INFO]:-For segments updated successfully,resynchronization will continue in the background.

20150602:22:16:57:062823gprecoverseg:mdw:gpadmin-[INFO]:-

20150602:22:16:57:062823gprecoverseg:mdw:gpadmin-[INFO]:-Use gpstate -s  to check theresynchronization progress.

20150602:22:16:57:062823gprecoverseg:mdw:gpadmin-[INFO]:-******************************************************************

 

4，--全量恢复，直接全部copy，才不管有什么变化。

[gpadmin@mdw ~]$ gprecoverseg -F

20150602:22:19:42:065360gprecoverseg:mdw:gpadmin-[INFO]:-Starting gprecoverseg with args: -F

20150602:22:19:42:065360gprecoverseg:mdw:gpadmin-[INFO]:-local Greenplum Version: 'postgres (GreenplumDatabase) 4.3.5.1 build 1'

20150602:22:19:42:065360gprecoverseg:mdw:gpadmin-[INFO]:-master Greenplum Version: 'PostgreSQL 8.2.15(Greenplum Database 4.3.5.1 build 1) on x86_64-unknown-linux-gnu, compiled byGCC gcc (GCC) 4.4.2 compiled on May 14 2015 14:07:14'

20150602:22:19:42:065360gprecoverseg:mdw:gpadmin-[INFO]:-Checking if segments are ready

20150602:22:19:42:065360gprecoverseg:mdw:gpadmin-[INFO]:-Obtaining Segment details from master...

20150602:22:19:43:065360gprecoverseg:mdw:gpadmin-[INFO]:-Obtaining Segment details from master...

20150602:22:19:43:065360gprecoverseg:mdw:gpadmin-[INFO]:-Greenplum instance recovery parameters

20150602:22:19:43:065360gprecoverseg:mdw:gpadmin-[INFO]:----------------------------------------------------------

20150602:22:19:43:065360gprecoverseg:mdw:gpadmin-[INFO]:-Recovery type              = Standard

20150602:22:19:43:065360gprecoverseg:mdw:gpadmin-[INFO]:----------------------------------------------------------

20150602:22:19:43:065360gprecoverseg:mdw:gpadmin-[INFO]:-Recovery 1 of 1

20150602:22:19:43:065360gprecoverseg:mdw:gpadmin-[INFO]:----------------------------------------------------------

20150602:22:19:43:065360gprecoverseg:mdw:gpadmin-[INFO]:-  Synchronization mode                        = Full

20150602:22:19:43:065360gprecoverseg:mdw:gpadmin-[INFO]:-  Failed instance host                        = sdw1

20150602:22:19:43:065360gprecoverseg:mdw:gpadmin-[INFO]:-  Failed instance address                     = sdw1

20150602:22:19:43:065360gprecoverseg:mdw:gpadmin-[INFO]:-  Failed instance directory                   = /data1/mirror/gpseg1

20150602:22:19:43:065360gprecoverseg:mdw:gpadmin-[INFO]:-  Failed instance port                        = 50000

20150602:22:19:43:065360gprecoverseg:mdw:gpadmin-[INFO]:-  Failed instance replication port            = 51000

20150602:22:19:43:065360gprecoverseg:mdw:gpadmin-[INFO]:-  Recovery Source instance host               = sdw2

20150602:22:19:43:065360gprecoverseg:mdw:gpadmin-[INFO]:-  Recovery Source instance address            = sdw2

20150602:22:19:43:065360gprecoverseg:mdw:gpadmin-[INFO]:-  Recovery Source instance directory          = /data1/primary/gpseg1

20150602:22:19:43:065360gprecoverseg:mdw:gpadmin-[INFO]:-  Recovery Source instance port              = 40000

20150602:22:19:43:065360gprecoverseg:mdw:gpadmin-[INFO]:-  Recovery Source instance replication port   = 41000

20150602:22:19:43:065360gprecoverseg:mdw:gpadmin-[INFO]:-  Recovery Target                             = in-place

20150602:22:19:43:065360gprecoverseg:mdw:gpadmin-[INFO]:----------------------------------------------------------

 

Continue withsegment recovery procedure Yy|Nn (default=N):

> y

20150602:22:19:45:065360gprecoverseg:mdw:gpadmin-[INFO]:-1 segment(s) to recover

20150602:22:19:45:065360gprecoverseg:mdw:gpadmin-[INFO]:-Ensuring 1 failed segment(s) are stopped

20150602:22:19:45:065360gprecoverseg:mdw:gpadmin-[INFO]:-14885: /data1/mirror/gpseg1

 

20150602:22:19:47:065360gprecoverseg:mdw:gpadmin-[INFO]:-Ensuring that shared memory is cleaned up forstopped segments

20150602:22:19:53:065360gprecoverseg:mdw:gpadmin-[INFO]:-Cleaning files from 1 segment(s)

.

20150602:22:19:54:065360gprecoverseg:mdw:gpadmin-[INFO]:-Building template directory

20150602:22:19:54:065360gprecoverseg:mdw:gpadmin-[INFO]:-Validating remote directories

.

20150602:22:19:55:065360gprecoverseg:mdw:gpadmin-[INFO]:-Copying template directory file

.

20150602:22:19:56:065360gprecoverseg:mdw:gpadmin-[INFO]:-Configuring new segments

.

20150602:22:19:57:065360gprecoverseg:mdw:gpadmin-[INFO]:-Cleaning files

.

20150602:22:19:58:065360gprecoverseg:mdw:gpadmin-[INFO]:-Starting file move procedure forsdw1:/data1/mirror/gpseg1:content=1:dbid=5:mode=r:status=d

updating flat files

20150602:22:19:58:065360gprecoverseg:mdw:gpadmin-[INFO]:-Updating configuration with new mirrors

20150602:22:19:58:065360gprecoverseg:mdw:gpadmin-[INFO]:-Updating mirrors

.

20150602:22:19:59:065360gprecoverseg:mdw:gpadmin-[INFO]:-Starting mirrors

20150602:22:19:59:065360gprecoverseg:mdw:gpadmin-[INFO]:-Commencing parallel primary and mirror segmentinstance startup, please wait.....

20150602:22:20:01:065360gprecoverseg:mdw:gpadmin-[INFO]:-Process results...

20150602:22:20:01:065360gprecoverseg:mdw:gpadmin-[INFO]:-Updating configuration to mark mirrors up

20150602:22:20:01:065360gprecoverseg:mdw:gpadmin-[INFO]:-Updating primaries

20150602:22:20:01:065360gprecoverseg:mdw:gpadmin-[INFO]:-Commencing parallel primary conversion of 1segments, please wait........

20150602:22:20:06:065360gprecoverseg:mdw:gpadmin-[INFO]:-Process results...

20150602:22:20:06:065360gprecoverseg:mdw:gpadmin-[INFO]:-Done updating primaries

20150602:22:20:06:065360gprecoverseg:mdw:gpadmin-[INFO]:-******************************************************************

20150602:22:20:06:065360gprecoverseg:mdw:gpadmin-[INFO]:-Updating segments for resynchronization iscompleted.

20150602:22:20:06:065360gprecoverseg:mdw:gpadmin-[INFO]:-For segments updated successfully,resynchronization will continue in the background.

20150602:22:20:06:065360gprecoverseg:mdw:gpadmin-[INFO]:-

20150602:22:20:06:065360gprecoverseg:mdw:gpadmin-[INFO]:-Use gpstate -s  to check theresynchronization progress.

20150602:22:20:06:065360gprecoverseg:mdw:gpadmin-[INFO]:-******************************************************************

[gpadmin@mdw ~]$gpstate -e

20150602:22:22:00:067423gpstate:mdw:gpadmin-[INFO]:-Starting gpstate with args: -e

20150602:22:22:00:067423gpstate:mdw:gpadmin-[INFO]:-local Greenplum Version: 'postgres (GreenplumDatabase) 4.3.5.1 build 1'

20150602:22:22:00:067423gpstate:mdw:gpadmin-[INFO]:-master Greenplum Version: 'PostgreSQL 8.2.15(Greenplum Database 4.3.5.1 build 1) on x86_64-unknown-linux-gnu, compiled byGCC gcc (GCC) 4.4.2 compiled on May 14 2015 14:07:14'

20150602:22:22:00:067423gpstate:mdw:gpadmin-[INFO]:-Obtaining Segment details from master...

20150602:22:22:00:067423gpstate:mdw:gpadmin-[INFO]:-Gathering data from segments...

.

20150602:22:22:01:067423gpstate:mdw:gpadmin-[INFO]:-----------------------------------------------------

20150602:22:22:01:067423gpstate:mdw:gpadmin-[INFO]:-Segment Mirroring Status Report

20150602:22:22:01:067423gpstate:mdw:gpadmin-[INFO]:-----------------------------------------------------

20150602:22:22:01:067423gpstate:mdw:gpadmin-[INFO]:-Segment Pairs in Resynchronization

20150602:22:22:01:067423gpstate:mdw:gpadmin-[INFO]:-   CurrentPrimary   Port    Resync mode   Est. resync progress   Total resync objects   Objects to resync   Data synced  Est. total to sync   Est. resyncend time   Change tracking size   Mirror  Port

20150602:22:22:01:067423gpstate:mdw:gpadmin-[INFO]:-   sdw2              40000   Full         99.50%                 2321                   0                   933 MB        938 MB               2015-06-02 22:22:00    150 MB                 sdw1     50000

 

--恢复时系统表的状态

testDB=# select *from gp_segment_configuration;

 dbid | content | role | preferred_role | mode| status | port  | hostname | address |replication_port | san_mounts

------+---------+------+----------------+------+--------+-------+----------+---------+------------------+------------

    2 |      0 | p    | p              | s    | u     | 40000 | sdw1     | sdw1    |           41000 |

    4 |      0 | m    | m              | s    | u     | 50000 | sdw2     | sdw2    |           51000 |

    3 |      1 | p    | p              | r    | u     | 40000 | sdw2     | sdw2    |           41000 |

    5 |      1 | m    | m              | r    | u     | 50000 | sdw1     | sdw1    |           51000 |

    1 |     -1 | p    | p              | s    | u     |  5432 | mdw      | mdw    |                  |

(5 rows)

 

--恢复成功

[gpadmin@mdw ~]$gpstate -e

20150602:22:23:51:068918gpstate:mdw:gpadmin-[INFO]:-----------------------------------------------------

20150602:22:23:51:068918gpstate:mdw:gpadmin-[INFO]:-Segment Mirroring Status Report

20150602:22:23:51:068918gpstate:mdw:gpadmin-[INFO]:-----------------------------------------------------

20150602:22:23:51:068918gpstate:mdw:gpadmin-[INFO]:-All segments are running normally

 

 

--恢复成功后系统表的状态

testDB=# select *from gp_segment_configuration;

 dbid | content | role | preferred_role | mode| status | port  | hostname | address |replication_port | san_mounts

------+---------+------+----------------+------+--------+-------+----------+---------+------------------+------------

    2 |      0 | p    | p              | s    | u     | 40000 | sdw1     | sdw1    |           41000 |

    4 |      0 | m    | m              | s   | u      | 50000 | sdw2     | sdw2   |            51000 |

    3 |      1 | p    | p              | s    | u     | 40000 | sdw2     | sdw2    |           41000 |

    5 |      1 | m    | m              | s    | u     | 50000 | sdw1     | sdw1   |            51000 |

    1 |     -1 | p    | p              | s    | u     |  5432 | mdw      | mdw    |                  |


--转载请注明出处blog.csdn.net/aabc012

D示羊
关注
  • 0
    点赞
  • 踩
  • 1
    收藏
    觉得还不错? 一键收藏
  • 1
    评论
专栏目录
GPDB gprecoverseg -F Unable to connect to database
数据人
04-27 1416
在GPDB运维过程中，有时会发生节点错误，一般采用gprecoverseg 进行节点的恢复，如果不能恢复，会尝试使用gprecoverseg -F进行全量恢复，一般在这个环节问题可以解决。但是个别情况下，会报出Unable to connect to database的错误，这个时候就需要特别处理了。 错误信息：报错信息为Unable to connect to database. Retryin
Greenplum segment节点异常恢复
无锋剑
08-30 2533
Segment检测及故障切换机制 GP Master首先会检测Primary状态，如果Primary不可连通，那么将会检测Mirror状态，Primary/Mirror状态总共有4种： Primary活着，Mirror活着。GP Master探测Primary成功之后直接返回，进行下一个Segment检测； Primary活着，Mirror挂了。GP Master探测Primary成功之后，通...
GreenPlum的那些事《二》——GreenPlun启动报错
热门推荐
数据一哥，公众号：数据社
05-15 1万+
标签（空格分隔）： GreenPlum 在配置好GreenPlum过程中，需要注意一下几点： 找好安装包 安装依赖，Gun版本需要注意 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 paral
greenplum全量恢复gprecoverseg -F出现Unable to connect to database时的相关分析及解决方法
aabc012的博客
07-28 1万+
之前有两位朋友碰到过在对greenplum的系统构架更改后，出现全量恢复gprecoverseg -F也无法正常运行的情况。 报错信息为Unable to connect to database. Retrying 1 gprecoverseg failed. (Reason='Unable to connect to database and start transaction') exiting... 有幸拷得一份虚拟机上的全部文件，对其进行分析。 发现其实出现这个问题只需要修改pg_changetra
gprecoverseg命令无法恢复Greenplum集群中挂掉的Segment实例
最新发布
Georgetwo的博客
12-09 1090
gprecoverseg命令无法恢复Greenplum集群中挂掉的Segment实例:Did not start the following segments due to failure while updating the port.Please update the port in postgresql.conf located in the segment's data directory
GreenPlum 大数据平台--segment 失效问题恢复《二》（全部segment宕机情况下）
Z14523299999的博客
07-11 319
01，情况描述 　　主Segment和它的镜像都宕掉。导致了greenplum数据库不可用状态 02，重启greenplum数据库 gpstop -r 03，恢复 gprecoverseg 04，状态检查 gpstate -m 保证所有segment处于Synchronized 05，存在处于Change Tracking模式 运行 ...
Postgresql备份和增量恢复方案
09-09
在设置增量备份时，首先要确保`wal_level`设置为`replica`或更高，这样会记录足够的信息以允许增量恢复。同时，启用`archive_mode`，并配置`archive_command`，以自动归档WAL日志。例如，可以设置`archive_command`...
mysql增量备份及断点恢复脚本实例
09-09
- 定期执行全备份，以简化恢复流程并减少依赖于多个增量备份的情况。 - 在恢复过程中，必须按照正确顺序应用全备份和增量备份，否则可能会导致数据不一致。 - 确保有适当的权限和安全措施，以防止未经授权的访问或误...
Mysql实现增量恢复的方法详解
09-09
MySQL增量恢复是数据库管理系统中的一种重要技术，用于在数据损坏或丢失的情况下恢复到某一特定时间点的状态。在本文中，我们将深入探讨MySQL增量恢复的原理、应用场景和具体实施步骤。 一、增量恢复适用的场景 1....
Oracle-RMAN增量备份恢复测试记录
12-28
Oracle-RMAN增量备份恢复测试记录 Oracle-RMAN增量备份恢复测试记录是指使用Oracle Recovery Manager（RMAN）工具对Oracle数据库进行增量备份和恢复的测试记录。本记录涵盖了增量备份的基本概念、备份策略及恢复、...
详细讲解西软FOXHIS增量备份与恢复方法
09-15
### 西软FOXHIS增量备份与恢复方法详解 #### 引言 在现代酒店管理信息系统中，数据安全至关重要。西软FOXHIS作为一款广泛应用于酒店业的信息管理系统，其数据的安全性和可靠性直接影响到酒店的正常运营和服务质量。...
greenplum 报错 valid segments to start the array
无锋剑
09-05 1740
greenplum 集群启动报错 Do not have enough valid segments to start the array. 前提： 集群配置完成后，有些集群配置需要优化调整一下： 设置work_mem 64MB 查看配置 gpconfig -s work_mem Values on all segments are consistent GUC : work...
Greenplum数据库配置管理-参数配置管理和常用参数优化建议
murkey的专栏
04-27 2247
GPDB 的系统配置 服务器配置文件包含着配置服务器行为的参数。Greenplum数据库的配置文件postgresql.conf 位于数据库实例的数据目录之下。 Master和每一个Segment实例都有自己的postgresql.conf文件。 一些参数是本地的：每个Segment实例检查它的postgresql.conf文件来得到这类 参数的值。在Master和每一个Segment实例上...
greenplum的主机名更改导致无法启动问题解决
qq_33445829的博客
02-24 678
查看报错日志 [gpadmin@mydb_master ~]$ cat /home/gpadmin/gpAdminLogs/gpstart_20210224.log 20210224:18:27:05:007350 gpstart:mydb_master:gpadmin-[INFO]:----------------------------------------------------- 20210224:18:27:05:007350 gpstart:mydb_master:gpadmin-[INFO
PostgreSQL服务器管理：服务器配置
weixin_33716941的博客
08-21 761
本文档为PostgreSQL 9.6.0文档，本转载已得到原译者彭煜玮授权。 1. 设置参数 1.1. 参数名称和值 所有参数名都是大小写不敏感的。每个参数都可以接受五种类型之一的值： 布尔、字符串、整数、 浮点数或枚举。该类型决定了设置该参数的语法： 布尔: 值可以被写成 on, off, true, false, yes, no, 1, 0 （都是...
greenplum一次节点迁移
wxc20062006的专栏
11-24 1761
greenplum节点迁移
数据库PostrageSQL-开发者选项
逍遥云恋
11-12 559
19.17. 开发者选项 下面的参数目的是用在PostgreSQL源代码上， 并且在某些情况下可以帮助恢复严重损坏了的数据库。在一个生产数据库中没有理由使用它们。同样，它们被从例子postgresql.conf文件中排除。请注意许多这些参数要求特殊的源代码编译标志才能工作。 allow_system_table_mods (boolean) 允许对系统表结构的修改。它可以被initdb使用。这个参数只能在服务器启动时设置。 ignore_system_indexes (boolean) 读取系统表时忽略系统
greenplum segment恢复的过程
朝着梦想 渐行前进
10-27 5615
greenplum segment 恢复
greenplum如何删除segment节点
DreamDB的专栏
04-20 2898
环境信息： Red Hat Enterprise Linux Server release 7.5 (Maipo) Greenplum Database 5.16.0 [gpadmin@rhmdw gpadata]$ cat /etc/hosts 127.0.0.1 localhost localhost.localdomain localhost4 localhost4.localdo...
Oracle RMAN增量备份与恢复实战记录
Oracle-RMAN增量备份恢复测试记录是一篇详细的文档，由朱晓凯在2012年3月29日创建，针对Oracle数据库管理工具RMAN（Recovery Manager）进行了增量备份和恢复操作的记录。本文档主要分为三个部分：一、概述了增量备份...
评论 1
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值