转 RMAN-20033

.备份日志里报错RMAN-20033：

RMAN-03008: error while performing automatic resync of recovery catalog

RMAN-20033: control file SEQUENCE# too low

最可原因是一个全备和一个归档日志备份同时进行，

Catalog 一次更新一次，导致另一个备份报错归档日志的序列号

过旧，

官方建议：是在catalog库重新注册一个db,或者重新初始化controlfile

Wednesday, 9 August 2017

RMAN-20033: control file SEQUENCE# too low

Most probably the true reason for that error to occur is a race condition between 2 concurrent backup processes - for example one full backup and one archivelog backup with autobackup enabled. Thus both would synchronize with RMAN catalog and one of them may try to register controlfile being older than the last one already registered by the 2nd process.

On Metalink they suggests a weird solution for this (Rman-20033 Control File Sequence# Too Low (Doc ID 342696.1)) to recreate the controlfile or RMAN catalog - it is possible it may be needed in some scenario, but I guess the best way first is simply rerun the backup ensuring there is only one backup process running.

解决办法：

重新注册db到catlog catalog

-首先检查catatlog 库，nods 信息

sqlplus rman11g/rman11g@cat11g

col name for a6

select * from rc_database where name='NODS';

DB_KEY DBINC_KEY DBID NAME RESETLOGS_CHANGE# RESETLOGS_TIME

---------- ---------- ---------- ------ ----------------- ------------------

26369 45331854 1522978174 NODS 1.3129E+10 17-SEP-17

102280754 102280755 1747437707 NODS 1 07-DEC-19

--unregister 跑一分钟

rman target / catalog rman11g/rman11g@cat11g

RMAN> unregister database;

database name is "NODS" and DBID is 1747437707

Do you really want to unregister the database (enter YES or NO)? YES

database unregistered from the recovery catalog

--如果遇到hang,可以打开pl/sql developer,看看host为odsdb的 session 等待是否在

等待，一般情况下不会有锁，只会有等待。

--如果等待是正常网络sqlnet等待，current sql那一栏没有显示数据，可以看看cursor游标那一栏，游标那一栏越往下，越是最新的

执行sql.

--也有可能不是hang,只是有可能一会是active,一会是inactive状态。