OGG 有兩天由於某種原因沒有啟動,而這段時間的備份文件缺失了一部分歸檔。恢復過程記錄如下:
GGSCI (xxxx) 6> info all Program Status Group Lag at Chkpt Time Since Chkpt MANAGER RUNNING EXTRACT RUNNING DP_XXXX3 00:00:00 00:00:08 EXTRACT RUNNING DP_XXXX4 00:00:00 00:00:05 EXTRACT ABENDED EX_XXXX3 00:00:01 65:09:29 EXTRACT ABENDED EX_XXXX4 00:00:01 65:09:35 GGSCI (xxxx) 6> view report EX_XXXX1 2019-05-27 14:58:34 ERROR OGG-00446 Could not find archived log for sequence 76437 thread 1 under default destinations SQL <SELECT name FROM v$archived_log WHERE sequence# = :ora_seq_no AND thread# = :ora_thread AND resetlogs_id = :ora_resetlog_id AND archived = 'YES' AND deleted = 'NO' AND standby_dest = 'NO' order by name DESC>, error retrievin g redo file name for sequence 76437, archived = 1, use_alternate = 0 Not able to establish initial position for sequence 76437, rba 6905872. 2019-05-27 14:58:34 ERROR OGG-01668 PROCESS ABENDING. 檢查確認: set line 200 pages 200 col name for a100 SELECT sequence#,name,thread#,resetlogs_id,archived,deleted FROM v$archived_log WHERE sequence# >= 76437; SEQUENCE# NAME THREAD# RESETLOGS_ID ARC DEL ---------- ---------------------------------------------------------------------------------------------------- ---------- ------------ --- --- 76437 1 898097516 YES YES 76438 1 898097516 YES YES 76439 1 898097516 YES YES 76440 1 898097516 YES YES 76441 1 898097516 YES YES 76442 1 898097516 YES YES 76443 1 898097516 YES YES 76444 1 898097516 YES YES 76445 1 898097516 YES YES 76446 1 898097516 YES YES 76447 1 898097516 YES YES 76448 1 898097516 YES YES 76449 1 898097516 YES YES 76450 1 898097516 YES YES 76451 1 898097516 YES YES 76452 1 898097516 YES YES 76453 1 898097516 YES YES 76454 1 898097516 YES YES 76455 1 898097516 YES YES 76456 1 898097516 YES YES 76457 1 898097516 YES YES 76458 1 898097516 YES YES 76459 1 898097516 YES YES 76460 1 898097516 YES YES 76461 1 898097516 YES YES 76462 1 898097516 YES YES 76463 1 898097516 YES YES 76464 1 898097516 YES YES 76465 1 898097516 YES YES 29 rows selected.
看不到歸檔路徑,DELETED='YES' ,可見確實歸檔丟失。
如果歸檔沒有丟失,將歸檔恢復到原來位置,重新啟動抽取進程,即可完成恢復。大概流程如下:
--1.檢查備份歸檔是否存在 rman target / RMAN > list backup of archivelog sequence between 76437 and 76501; --2.恢復歸檔 RMAN > restore archivelog from sequence 76437; --3.確認歸檔恢復到了默認歸檔位置 --4.重新啟動抽取進程 ./ggsci start ex_xxxx3 info all stats ex_xxxx3 stats dp_xxxx3 比對數據,確認恢復
再次確認后,RAC 兩個節點的歸檔都有丟失。
thread2: 52036 到 52041斷檔
thread1: 76495 到 76501斷檔
其中
進程3 對應的表只有insert操作,我們采用如下方式處理的:
比對數據將這段時間的數據插入進去(可以采用tns/exp where/plsql where/expdp where等方式),然后跳過這段時間進行同步
進程4 包含了更新插入刪除操作,只能通過重新初始化的方式恢復
進程3 補數據恢復
--1.獲取斷檔故障時間范圍 SQL> alter session set nls_date_format='YYYYMMDD HH24:MI:ss'; Session altered. SQL> select THREAD#,SEQUENCE#,FIRST_TIME,NEXT_TIME from v$archived_log where SEQUENCE# between 52036 and 52041; SEQUENCE# FIRST_TIME NEXT_TIME ---------- ----------------- ----------------- 52036 20190526 20:05:42 20190526 21:05:41 52037 20190526 21:05:41 20190526 21:34:56 52038 20190526 21:34:56 20190526 22:34:54 52039 20190526 22:34:54 20190526 23:34:54 52040 20190526 23:34:54 20190527 00:34:55 52041 20190527 00:34:55 20190527 01:34:55 6 rows selected. SQL> select THREAD#,SEQUENCE#,FIRST_TIME,NEXT_TIME from v$archived_log where SEQUENCE# between 76495 and 76501; THREAD# SEQUENCE# FIRST_TIME NEXT_TIME ---------- ---------- ----------------- ----------------- 1 76495 20190526 20:05:29 20190526 20:05:42 1 76496 20190526 20:05:42 20190526 20:56:07 1 76497 20190526 20:56:07 20190526 21:34:54 1 76498 20190526 21:34:54 20190526 22:24:17 1 76499 20190526 22:24:17 20190526 23:24:15 1 76500 20190526 23:24:15 20190527 00:24:15 1 76501 20190527 00:24:15 20190527 01:24:17 --2.將這段時間數據插入過去 insert into target_table select * from source_table where createtime>=to_date('20190526 20:05:29','YYYYMMDD HH24:MI:SS') AND createtime<=to_date('20190527 01:34:55','YYYYMMDD HH24:MI:SS'); --3.跳過這段數據同步 GGSCI (xxxx) 2> ALTER EXTRACT EX_XXXX3, THREAD 2, BEGIN 2019-05-27 01:34:56 EXTRACT altered. GGSCI (xxxx) 2> ALTER EXTRACT EX_XXXX3, THREAD 1, BEGIN 2019-05-27 01:24:18 EXTRACT altered. GGSCI (xxxx) 14> ALTER EXTRACT EX_XXXX3, ETROLLOVER 2019-05-28 17:21:15 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 (xxxx) 4> start EX_XXXX3 Sending START request to MANAGER ... EXTRACT EX_XXXX3 starting GGSCI (xxxx) 4> info all --檢查狀態 GGSCI (xxxx) 4> info exttrail ./dirdat/lc Extract Trail: ./dirdat/lc Extract: EX_XXXX3 Seqno: 2831 --tracefile SEQNO已經跳轉到2831 RBA: 78985035 File Size: 100M GGSCI (xxxx) 7> info DP_XXXX3 EXTRACT DP_XXXX3 Last Started 2019-05-27 14:47 Status RUNNING Checkpoint Lag 00:00:00 (updated 00:00:04 ago) Log Read Checkpoint File ./dirdat/lc002830 --DB進程的trace文件不能自動跳轉 2019-05-28 17:12:21.854829 RBA 1084
GGSCI (xxxx) 7> ALTER EXTRACT DP_XXXX3, EXTSEQNO 2831, EXTRBA 0
GGSCI (xxxx) 7> START DP_XXXX3
目標端
GGSCI (xxxx) 1> start RP_XXXX3
GGSCI (xxxx) 1> stats RP_XXXX3 --檢查是否有數據,完成恢復
進程4 重新初始化恢復
源端:
1.根據SCN 導出數據
sqlplus ggadmin/ggadmin
select to_char(current_scn) from v$database; --15555721660
expdp userid=scott/tiger directory=expdir tables=testtable1 dumpfile=testtable1.dmp logfile=testtable1.log cluster=n CONTENT=data_only FLASHBACK_SCN=15555721660
scp 到目標端
2.目標端導入
impdp target/target directory=expdir remap_schema=scott:target dumpfile=testtable1.dmp table_exists_action=truncate
3. 修復OGG
隨便找個不斷檔的時間點恢復即可
STOP EX_XXXX4 ALTER EXTRACT EX_XXXX4, THREAD 2, BEGIN 2019-05-28 16:00:00 ALTER EXTRACT EX_XXXX4, THREAD 1, BEGIN 2019-05-28 16:00:00 ALTER EXTRACT EX_XXXX4, ETROLLOVER 2019-05-28 17:21:15 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. start EX_XXXX4 info all GGSCI (xxxx) 4> info exttrail ./dirdat/ld Extract Trail: ./dirdat/lc Extract: EX_XXXX3 Seqno: 2638 RBA: 45765133 File Size: 100M STOP DP_XXXX4 ALTER EXTRACT DP_XXXX4, EXTSEQNO 2638, EXTRBA 0 START DP_XXXX4 目標端: START REPLICAT RP_XXXX4,ATCSN 15555721660
完成恢復。
以上過程是后期梳理,並沒有保留詳細日志,所以存在一些出入,但大概操作步驟如此。