OGG 11g Checkpoint 詳解
2024-07-21 02:52:52
供稿：網(wǎng)友
OGG Checkpoint 詳解定位中斷的位置，下次啟動(dòng)從中斷的位置開始恢復(fù)。1.target 端配置；2.一條記錄對(duì)應(yīng)一個(gè)replicat一. Extract Checkpoints (1) Extract read checkpoints （1） Startup checkpoint --ogg啟動(dòng)的時(shí)候創(chuàng)建 當(dāng)Extract 進(jìn)程啟動(dòng)的時(shí)候，會(huì)創(chuàng)建startup checkpoint，用來記錄此時(shí)record的相關(guān)信息。該 checkpoint 是第一個(gè)checkpoint。 startup checkpoint的統(tǒng)計(jì)信息包含如下內(nèi)容： （1） Thread #: 創(chuàng)建checkpoint 的thread編號(hào)。 （2） Sequence #:創(chuàng)建checkpoint 時(shí)對(duì)應(yīng)的sequence 號(hào)碼。 --list archive log; （3） RBA: 創(chuàng)建checkpoint 時(shí)，redo log 里面record 的relative byte address。 （4）Timestamp: 創(chuàng)建checkpoint 時(shí)record對(duì)應(yīng)的timestamp。 （5）SCN: 創(chuàng)建checkpoint 時(shí) record 對(duì)應(yīng)的system change number 。 （6）Redo File: 創(chuàng)建checkpoint 時(shí)record對(duì)應(yīng)的事務(wù)日志路徑。 2 ) Recovery checkpoint recover checkpoint 記錄的是data source（online redo log 或者歸檔中）中Extract 最久沒處理的record的位置。 該checkpoint 的統(tǒng)計(jì)信息和startup checkpoint一樣。3) Current checkpointcurrent checkpoint 是data source中Extract 最后一次讀取的record的位置。GGSCI (ogg02) 4> info ext1EXTRACT EXT1 Last Started 2014-12-27 17:42 Status RUNNINGCheckpoint Lag 00:00:00 (updated 00:00:04 ago)PRocess ID 31621Log Read Checkpoint Oracle Redo Logs 2014-12-27 18:06:05 Seqno 15, RBA 18128896 SCN 0.1142344 (1142344)Seqno 和 RBA 會(huì)變，說明ogg是在工作的。GGSCI (ogg02) 5> info ext1,showchEXTRACT EXT1 Last Started 2014-12-27 17:42 Status RUNNINGCheckpoint Lag 00:00:00 (updated 00:00:00 ago)Process ID 31621Log Read Checkpoint Oracle Redo Logs 2014-12-27 18:11:06 Seqno 15, RBA 18188800 SCN 0.1142472 (1142472)Current Checkpoint Detail:Read Checkpoint #1 Oracle Redo Log Startup Checkpoint (starting position in the data source): Thread #: 1 Sequence #: 11 RBA: 7646224 Timestamp: 2014-12-27 16:20:48.000000 SCN: 0.1136079 (1136079) Redo File: /oradata/ogg02/redo02.log Recovery Checkpoint (position of oldest unprocessed transaction in the data source): Thread #: 1 Sequence #: 15 RBA: 18187792 Timestamp: 2014-12-27 18:09:41.000000 SCN: 0.1142443 (1142443) Redo File: /oradata/ogg02/redo03.log Current Checkpoint (position of last record read in the data source): Thread #: 1 Sequence #: 15 RBA: 18188800 Timestamp: 2014-12-27 18:11:06.000000 SCN: 0.1142472 (1142472) Redo File: /oradata/ogg02/redo03.logWrite Checkpoint #1 GGS Log Trail Current Checkpoint (current write position): Sequence #: 4 RBA: 1755 Timestamp: 2014-12-27 18:11:43.251745 Extract Trail: /u01/zt/dirdat/lt Trail Type: RMTTRAILHeader: Version = 2 Record Source = A Type = 10 # Input Checkpoints = 1 # Output Checkpoints = 1File Information: Block Size = 2048 Max Blocks = 100 Record Length = 2048 Current Offset = 0Configuration: Data Source = 3 Transaction Integrity = 1 Task Type = 0Status: Start Time = 2014-12-27 17:42:32 Last Update Time = 2014-12-27 18:11:43 Stop Status = A Last Result = 400二.Extract write checkpoints該checkpoint包含如下信息：（1）Sequence #: checkpoint 正在寫的trail 文件的序列號(hào)。（2）RBA: 創(chuàng)建checkpoint 時(shí)trail文件里record 的relative byte address。（3）Timestamp: 創(chuàng)建checkpoint 時(shí)record對(duì)應(yīng)的timestamp。（4）Extract trail: trail 文件的路徑。（5）Trail Type: 標(biāo)識(shí)trail 文件類型。有2個(gè)值：EXTTRAIL：表示該trail 是本地的trail，文件由OGG進(jìn)程直接寫如本地disk。RMTTRAIL：表示是遠(yuǎn)程的trail，該文件不會(huì)寫入本地的磁盤，而是傳到遠(yuǎn)端的的OGG上，在有遠(yuǎn)端OGG的collector 進(jìn)程寫disk。 比如pump 進(jìn)程，里面就會(huì)寫RMTTRAIL。-----------------------------------------------------------------------------------------target 端GGSCI (ogg04) 1> info rep1REPLICAT REP1 Last Started 2014-12-27 14:07 Status RUNNINGCheckpoint Lag 00:00:00 (updated 00:00:01 ago)Process ID 26088Log Read Checkpoint File /u01/zt/dirdat/lt000004 2014-12-27 16:49:03.408880 RBA 1755GGSCI (ogg04) 2> info rep1,showchREPLICAT REP1 Last Started 2014-12-27 14:07 Status RUNNINGCheckpoint Lag 00:00:00 (updated 00:00:07 ago)Process ID 26088Log Read Checkpoint File /u01/zt/dirdat/lt000004 2014-12-27 16:49:03.408880 RBA 1755Current Checkpoint Detail:Read Checkpoint #1 GGS Log Trail Startup Checkpoint (starting position in the data source): Sequence #: 3 RBA: 20259 Timestamp: 2014-12-27 14:06:57.000000 Extract Trail: /u01/zt/dirdat/lt Current Checkpoint (position of last record read in the data source): --最后一次讀取文件的位置 Sequence #: 4 RBA: 1755 Timestamp: 2014-12-27 16:49:03.408880 Extract Trail: /u01/zt/dirdat/ltHeader: Version = 2 Record Source = A Type = 1 # Input Checkpoints = 1 # Output Checkpoints = 0File Information: Block Size = 2048 Max Blocks = 100 Record Length = 2048 Current Offset = 0Configuration: Data Source = 0 Transaction Integrity = -1 Task Type = 0Database Checkpoint: Checkpoint table = ggs.checkpoint Key = 982417365 (0x3a8e7fd5) Create Time = 2014-12-27 10:46:19Status: Start Time = 2014-12-27 14:07:31 Last Update Time = 2014-12-27 17:03:30 Stop Status = A Last Result = 400 GGSCI (ogg04) 3> info rep1 --看rba是否有變化 REPLICAT REP1 Last Started 2014-12-27 14:07 Status RUNNINGCheckpoint Lag 00:00:00 (updated 00:00:08 ago)Process ID 26088Log Read Checkpoint File /us01/zt/dirdat/lt000004 2014-12-27 16:49:03.408880 RBA 1755----------------------------------------------------------------------停的時(shí)候確保兩邊都事物都提交了。 info 一下看看 ②. data dump進(jìn)程：將隊(duì)列數(shù)據(jù)發(fā)送到目標(biāo) 好處：① .預(yù)防網(wǎng)絡(luò)和源庫的故障； ②. 可以對(duì)數(shù)據(jù)進(jìn)行過來和轉(zhuǎn)換； ③.synchronize方式從多個(gè)源庫上合并數(shù)據(jù)到一個(gè)中心庫； ④. synchronize方式將一個(gè)源庫數(shù)據(jù)傳到多個(gè)目標(biāo)庫;轉(zhuǎn)：http://blog.csdn.net/zhengwei125/article/details/46503403
