goldengate从11.2升级到12.1.2
1.停止抽取进程
| GGSCI (001.oracle.drs.dc.com) 286> stop EXTSJ01 |
2. 停止投递和复制进程
等待投递进程传输完和复制进程加载完毕.当投递进程的Write Checkpoint #1的Sequence和RBA和复制进程的 Current CHECKPOINT的Sequence和RBA相同时表示已同步完毕
2.1检查投递进程：
| GGSCI (001.oracle.drs.dc.com) 202> info PUPTDB2P,showch
EXTRACT PUPTDB2P Last Started 2016-05-04 20:32 Status STOPPED Checkpoint Lag 00:00:00 (updated 01:12:12 ago) Log Read Checkpoint File ./dirdat/exttrdbk/ex000006 First Record RBA 1151
Current Checkpoint Detail:
Read Checkpoint #1
GGS Log Trail
Startup Checkpoint (starting position in the data source): Sequence #: 6 RBA: 1151 Timestamp: 2016-04-29 14:40:29.589038 Extract Trail: ./dirdat/exttrdbk/ex
Current Checkpoint (position of last record read in the data source): Sequence #: 6 RBA: 1151 Timestamp: Not Available Extract Trail: ./dirdat/exttrdbk/ex
Write Checkpoint #1
GGS Log Trail
Current Checkpoint (current write position): Sequence #: 7 RBA: 1183 Timestamp: 2016-05-04 20:35:56.803177 Extract Trail: /home/ggt/goldengate/dirdat/reptdb2p/re Trail Type: EXTTRAIL
|
2.2检查复制进程
| GGSCI (001.oracle.drs.dc.com) 209> info reptdb2p,showch
REPLICAT REPTDB2P Last Started 2016-05-04 20:32 Status STOPPED Checkpoint Lag 00:00:00 (updated 01:14:28 ago) Log Read Checkpoint File ./dirdat/reptdb2p/re000007 2016-05-04 20:32:43.741679 RBA 1183
Current Checkpoint Detail:
Read Checkpoint #1
GGS Log Trail
Startup Checkpoint (starting position in the data source): Sequence #: 6 RBA: 1246 Timestamp: 2016-04-29 14:40:29.589110 Extract Trail: ./dirdat/reptdb2p/re
Current Checkpoint (position of last record read in the data source): Sequence #: 7 RBA: 1183 Timestamp: 2016-05-04 20:32:43.741679 Extract Trail: ./dirdat/reptdb2p/re |
2.3停止投递进程和复制进程
| GGSCI (001.oracle.drs.dc.com) 209> STOP puptdb2p GGSCI (001.oracle.drs.dc.com) 210> STOP reptdb2p |
3.停止管理进程
| GGSCI (001.oracle.drs.dc.com) 210> STOP mgr |
4.备份原来的goldengate
| [ggt@001.oracle.drs.dc.com ~]$ mv goldengate goldengate11.2 |
5.安装goldengate 12.1
| [ggt@001 ~]$ unzip 121200_fbo_ggs_Linux_x64_shiphome.zip [ggt@001 ~]$ cd fbo_ggs_Linux_x64_shiphome/Disk1 [ggt@001 ~]$ ./runInstaller |
选择数据库版本，我的是oracle 11G
选择安装位置：指定到原始版本的位置；选择是否自动启动管理进程
6.检查安装有效性
| [ggt@001.oracle.drs.dc.com goldengate]$ ./ggsci
Oracle GoldenGate Command Interpreter for Oracle Version 12.1.2.0.0 17185003 OGGCORE_12.1.2.0.0_PLATFORMS_130924.1316_FBO Linux, x64, 64bit (optimized), Oracle 11g on Sep 25 2013 00:31:13 Operating system character set identified as UTF-8.
Copyright (C) 1995, 2013, Oracle and/or its affiliates. All rights reserved. |
7.把原版本的相关目录复制到新版本下
| [ggt@001 goldengate]$ cp -R /home/ggt/goldengate11.2/dirchk/ . cp -R /home/ggt/goldengate11.2/dirdat/ . cp -R /home/ggt/goldengate11.2/dirdef/ . cp -R /home/ggt/goldengate11.2/dirpcs/ . cp -R /home/ggt/goldengate11.2/dirprm/ . cp -R /home/ggt/goldengate11.2/dirrpt/ . cp -R /home/ggt/goldengate11.2/dirsql/ . cp -R /home/ggt/goldengate11.2/dirtmp/ . cp -R /home/ggt/goldengate11.2/BR/ . |
8.升级复制进程的checkpointtable
| GGSCI (001.oracle.drs.dc.com) 20> dblogin userid ggt@mpaysoa_kx, password pwd Successfully logged into database. GGSCI (001.oracle.drs.dc.com) 22> upgrade checkpointtable chk_mpayb2p
Successfully upgraded checkpoint table chk_mpayb2p.
注意：这一步必须做，否则这报以下错： 2016-05-05 16:31:56 ERROR OGG-00868 OCI Error ORA-00904: "LOG_BSN": invalid identifier (status = 904). Updating checkpoint table chk_mpay b2p, group REPMPB2P, key 169,822,320 (0x000000000a1f4870), SQL <UPDATE chk_mpayb2p SET last_update_ts = sysdate, seqno = :seqno, rba = :rba, audit_ts = :audit_ts, log_bsn = :log_bsn, log_csn = :log_csn, log_xid = :log_xid, log_cmplt_csn = :log_cmplt_csn, log_cmplt_xids = :log_cmplt _xids, version = :version WHERE group_name = :group_name AND group_key = :key>. |
9.启动管理进程
| GGSCI (001.oracle.drs.dc.com) 23>start mgr |
10.启动抽取进程
注意：启动之前，要修改trail文件的格式或把trail的序号向前滚一个序号,否则会报以下错：
ERROR OGG-01416 File ./dirdat/lt000005, with format RELEASE 11.2, does not match current format specification of RELEASE 12.1. Modify the parameter file to specify format RELEASE 11.2 or issue ETROLLOVER prior to restart.
方法一：修改trail文件的格式方法：把开参数文件，在exttrail ./dirdat/sj01/ex后面加上FORMAT RELEASE 11.2
| GGSCI (001.oracle.drs.dc.com) 23>edit params extsj01 修改exttrail ./dirdat/sj01/ex为exttrail ./dirdat/sj01/ex, FORMAT RELEASE 11.2 |
方法二：trail的序号向前滚一个序号
| GGSCI (001.oracle.drs.dc.com) 262> alter extsj01,etrollover
2016-05-05 10:14:51 INFO OGG-01520 Rollover performed. For each affected output trail of Version 10 or higher format, after starting the source extract, issue ALTER EXTSEQNO for that trail's reader (either pump EXTRACT or REPLICAT) to move the reader's scan to the new trail file; it will not happen automatically. EXTRACT altered. |
启动抽取进程
| GGSCI (001.oracle.drs.dc.com) 262>start extsj01 |
11.启动投递进程
如果在第10步用方法一的话，直接启动投递进程和复制进程就可以了，如果用方法二必须执行以下操作
11.1 记录投递进程的读写检查点
| GGSCI (001.oracle.drs.dc.com) 258> info PUPMPBK2,showch
EXTRACT PUPMPBK2 Last Started 2016-05-05 10:01 Status ABENDED Checkpoint Lag 00:44:46 (updated 00:00:21 ago) Log Read Checkpoint File /home/ggt/goldengate/dirdat/sj01/ex194468 2016-05-05 09:16:27.000000 RBA 2244115
Current Checkpoint Detail:
Read Checkpoint #1
GGS Log Trail
Startup Checkpoint (starting position in the data source): Sequence #: 194468 RBA: 2244115 Timestamp: 2016-05-05 09:16:27.000000 Extract Trail: /home/ggt/goldengate/dirdat/sj01/ex
Current Checkpoint (position of last record read in the data source): Sequence #: 194468 RBA: 2244115 Timestamp: 2016-05-05 09:16:27.000000 Extract Trail: /home/ggt/goldengate/dirdat/sj01/ex
Write Checkpoint #1
GGS Log Trail
Current Checkpoint (current write position): Sequence #: 6184 RBA: 36936583 Timestamp: 2016-05-05 09:16:30.784940 Extract Trail: /home/ggt/goldengate/dirdat/repmpaybak/re Trail Type: RMTTRAIL
Header: Version = 2 Record Source = A Type = 1 # Input Checkpoints = 1 # Output Checkpoints = 1
File Information: Block Size = 2048 Max Blocks = 100 Record Length = 2048 Current Offset = 0
Configuration: Data Source = 0 Transaction Integrity = 1 Task Type = 0
Status: Start Time = 2016-05-05 10:01:08 Last Update Time = 2016-05-05 10:01:13 Stop Status = A Last Result = 0 |
11.2 修改Read Checkpoint和Write Checkpoint
| GGSCI (001.oracle.drs.dc.com) 262> alter PUPMPBK2,etrollover
2016-05-05 10:14:51 INFO OGG-01520 Rollover performed. For each affected output trail of Version 10 or higher format, after starting the source extract, issue ALTER EXTSEQNO for that trail's reader (either pump EXTRACT or REPLICAT) to move the reader's scan to the new trail file; it will not happen automatically. EXTRACT altered.
GGSCI (001.oracle.drs.dc.com) 262> alter PUPMPBK2,extseqno 194469 extrba 0 注意：extseqno 194469是在Read Checkpoint的Sequence #: 194468加1，因为在抽取进程上用etrollover向前滚了一个序号
GGSCI (001.oracle.drs.dc.com) 262>start PUPMPBK2 |
11.3启动投递进程
| GGSCI (001.oracle.drs.dc.com) 262>start PUPMPBK2 |
12. 启动复制进程
12.1查看复制进程的加载位置
| GGSCI (001.oracle.drs.dc.com) 21> info REPMPB2P,showch
REPLICAT REPMPB2P Last Started 2016-05-05 16:31 Status ABENDED Checkpoint Lag 00:00:00 (updated 00:04:42 ago) Log Read Checkpoint File ./dirdat/repmpb2p/re000083 First Record RBA 0
Current Checkpoint Detail:
Read Checkpoint #1
GGS Log Trail
Startup Checkpoint (starting position in the data source): Sequence #: 6184 RBA: 36936583 Timestamp: Not Available Extract Trail: ./dirdat/repmpb2p/re
Current Checkpoint (position of last record read in the data source): Sequence #: 6184 RBA: 36936583 Timestamp: Not Available Extract Trail: ./dirdat/repmpb2p/re |
12.2修改复制进程的Sequence #和RBA
| GGSCI (001.oracle.drs.dc.com) 22>alter REPMPB2P extseqno 6185 extrba 0 注意：extseqno 6185是在Current Checkpoint的Sequence #: 6184加1，因为在投递进程上用etrollover向前滚了一个序号
|
12.3 启动复制进程
| GGSCI (001.oracle.drs.dc.com) 22>start REPMPB2P |
13 升级完成
恭喜你，至此你已成功完成goldengate的11g到12c的升级