RMAN backup compression and tape hardware compression

Hello,
Should tape backup compression be ON and RMAN compression OFF? What is the correct Oracle answer?
Common knowledge says that if both hardware and software compression were allowed to be enabled at the same time the resulting file in some cases may actually be bigger than the original file. Hardware data compression is done by specialized chips and faster than software compression and increases the tape performance.
I was reading somewhere some time ago that tape hardware will actually not attempt to compress data if the the data being processed cannot be compressed any further. But even if this is a hoax, wouldn't rman compression, at the cost of CPU, save I/O bandwidth?
Thanks.
Edited by: MaC on Dec 19, 2010 4:00 PM

MaC wrote:
I just located Oracle document ID 1018242.1, which outlines:
Use either hardware or software compression, but NOT both. Using both will actually increase the size of the resulting file. The rule of thumb is: compression + compression = no_compression.
Sounds kind a logic.I was actually going to post the same sentence but you did find it on your own :) . Another thing which is worth to mention that from 11.2, the RMAN backup compression is licensed with 4 options, High, Medium and Low . These all three options requires you to have Advanced Compression License to buy. There is another 4th option, Default which is free from license and the compression ratio of that lies in between of Medium and Low.
HTH
Aman....

Similar Messages

NEW 9I RMAN: BACKUP OPTIMIZATION AND RETENTION POLICIES

제품 : RMAN
작성날짜 : 2004-05-20
NEW 9I RMAN: BACKUP OPTIMIZATION AND RETENTION POLICIES
=======================================================
오라클 9i RMAN 에서는 백업을 보다 효율적으로 할 수 있는 기능을 보강 했다.
지금 소개할 "Retention Policies" 기능을 통해서 보다 효율적인 RMAN 백업
전략을 세울 수 있다.
1. 백업 화일 최적화 방법 (Backup File Optimization)
백업 화일 최적화란 백업에 소요되는 공간을 최소화 하는 전략이다. RMAN
백업시에 같은 정보 (dbid, checkpoint, and resetlogs data 등등) 를 가지고 있는
화일이 이미 존재하는 백업 셋 내부에 있는지 체크하게 된다. 만일 같은 화일이
이미 백업 되어 있다면 BACK UP 명령은 해당 화일에 대해서는 백업을 하지 않게
된다.
이때 같은 화일이라고 판단하는 기준은 다음과 같다.
* Datafile: 같은 DBID, checkpoint SCN, resetlogs SCN 과 time. 데이타 화일은
반드시정상적으로 offline 되었거나, read-only 이거나, 또는 정상적 으로 close
되어야 함.
* Archived redo log: 같은 thread, sequence number, 그리고 같은 Resetlogs
SCN 과 time.
* Backup Set: 같은 Backup Set recid 와 stamp.
RMAN 이 백업을 수행하다가 위와 같은 조건의 화일이 이미 존재함을 확인하면
이것은 건너뛰게 될 화일의 대상이다. 하지만 이때 바로 Skip 을 결정하지 않고
정해진 Retention Policies 를 조사 한후에 Skip 여부를 결정 하게 된다.
만일 백업 명령에 DELETE INPUT option 이 사용되면 RMAN 은 백업이 Skip
되어도 화일을 지우게 된다.
RMAN 은 모든 화일에 대한 백업이 Skip 되어도 에러메시지나 경고를 보내지
않는다.
그러나 만일 데이터 화일에 대한 백업이 recovery policy window 보다 오래
된 것 이면 RMAN 은 새로운 백업 화일을 만들기 위해서 화일을 백업 하게 된다.
Note:
자체적인 expirations policy 를 가지고 있는 media manager 를 사용할 경우에는
이런 백업 최적화 정책을 사용하는 것에 신중을 기해야 한다.
백업 최적화 를 사용하기 위해서는 CONFIGURE 명령을 사용한다. CONFIGURE
명령은 명령 수행 이후에 이루어지는 모든 백업에 대해 적용이 된다.
예:
CONFIGURE BACKUP OPTIMIZATION ON; # default 는 OFF
2. Backup Optimization에 Retention Policies 적용 하기
Retention Policy 로 백업 최적화를 조절 할 수 있다. 그러나 retention policy를
사용하지 않기 위해서는 'CONFIGURE RETENTION POLICY TO NONE' 을
이용해서 명시적으로 retention policy를 사용하지 않는다고 해야 한다. 디폴트로
REDUNDANCY = 1 이 적용 된다.
a. Recovery Window 를 이용한 Backup Optimization
만일 백업 최적화가 enable 되어 있고 Recovery Window 가 retention policy
적용을 위해서 셋업 되어 있으면 RMAN 은 항상 가장 최근의 백업이 Recovery
Window 보다 오래된 데이터 화일을 백업한다.
예를 들면 다음과 같은 조건에서
o Today is February 21.
o The recovery window is 7 days.
o The most recent backup of tablespace tbs2 to tape is January 3.
o Tablespace tbs2 is read-only.
2월 21일에 tbs2 테이블 스페이스를 테이프로 백업을 하라는 명령을 내리면,
RMAN 은 이 화일이 1월3일 이후에 변경 사항이 없는데도 불구 하고 백업을 하게
된다. 이로써 RMAN 은 최근 7일 동안에 최소한 한번의 백업이 있어야 된다는
조건을 지키게 된다.
이러한 작동 방식은 media manager 가 오래된 테잎을 제거 해도 되도록 한다.
그렇지 않다면, media manager 는 1월 3일의 백업을 무한정 가지고 있게 된다.
2월 21일에 tbs2 테이블 스페이스의 보다 최근의 백업을 만듦으로 써, RMAN 은
media manager 가 1월 3일에 백업 했던 테입을 지워도 되도록 한다.
b. Redundancy 를 이용한 Backup Optimization
Retention policy 로 Redundancy 를 채택 한 경우에 RMAN 은 Redundancy
에 1을 더한 갯수를 초과하는 오프라인 또는 읽기 전용 화일의 백업을 건너뛴다.
이때 'CONFIGURE RETENTION POLICY TO REDUNDANCY n' 명령어로
Redundancy의 갯수를 정한다.
아래와 같은 백업 최적화 명령을 내린 경우를 예로 든다:
CONFIGURE BACKUP OPTIMIZATION ON;
CONFIGURE RETENTION POLICY TO REDUNDANCY 2;
아카이브 테이블스페이스를 한번도 백업을 받은적이 없고 다음의 작업을 1주일
동안 한다고 하자.
Day Action Result Redundant Backup
Monday Take tablespace archive offline clean.
Tuesday Run BACKUP DATABASE. The archive tablespace is backed up.
Wednesday Run BACKUP DATABASE. The archive tablespace is backed up.
Thursday Run BACKUP DATABASE. The archive tablespace is backed up. Tuesday backup.
Friday Run BACKUP DATABASE. n/a Tuesday backup.
Saturday Run BACKUP DATABASE. n/a Tuesday backup.
Sunday Run DELETE OBSOLETE. The Tuesday backup is deleted.
Monday Run BACKUP DATABASE. The archive tablespace is backed up. Wednesday backup.
화요일, 수요일, 그리고 목요일의 백업은 아카이브 테이블 스페이스를 복사 해서
3개의 백업이 반드시 존재 해야 한다는 조건을 충촉 시킨다. (1+Redundancy)
금요일과 토요일에는 백업 최적화 조건에 의해서 아카이브 테이블 스페이스를
복사하지 않게 된다.
RMAN 은 일요일에는 유효기간이 지난 백업 화일을 지우게 된다. 따라서 화요일에
만든 백업은 삭제 된다. 월요일의 전체 백업은 3개의 백업이 존재 해야 한다는 조건
때문에 아카이브 테이블 스페이스를 또 다시 백업하게 된다. 이런 방식으로 백업
사이클이 진행 된다.
3. RETENTION POLICY NONE vs. DEFAULT
주의 사항:
'CONFIGURE RETENTION POLICY TO NONE' 과 'CONFIGURE RETENTION
POLICY TO DEFAULT' 는 그 의미가 같지 않다. 전자는 RETENTION POLICY
자체가 존재 하지 않는다는 의미 이며 백업은 expire 되지 않으며 'DELETE
OBSOLETE' 명령은 사용 할 수 없게 된다. 후자는 디폴트 RETENTION POLICY
(REDUNDANCY 1) 를 사용하게 된다는 의미 이다.
'DELETE OBSOLETE' 명령은 RETENTION POLICY 기준으로 expire 된 백업을
제거 하라는 명령이다.
보다 자세한 내용은 Oracle9i Recovery Manager User's Guide and
Reference의 Backup Optimization 부분을 참고 하시기 바랍니다.
--------------THE END----------------------------------------------

Hi,
Backup optimisation = +- do not copy empty space.
So, when you issue your BACKUP statement, it backs up the datafile. dot. nothing more.
In order to achieve what you're testing, look into TFM for BACKUP INCREMENTAL LEVEL x .
Regards,
Yoann.

The detail algorithm of OLTP table compress and basic table compress?

I'm doing a research on the detail algorithm of OLTP table compress and basic table compress, anyone who knows, please tell me. 3Q, and also the difference between them

http://www.oracle.com/us/products/database/db-advanced-compression-option-1525064.pdf
Edited by: Sanjaya Balasuriya on Dec 5, 2012 2:49 PM
Edited by: Sanjaya Balasuriya on Dec 5, 2012 2:49 PM

RMAN Backup Restore from tape

Hello Gurus
We have lvl0 RMAN database backup restored from tape but due to some reasons we could not apply recovery on it. On checking the information in controlfile, it identified only the system file and no other file. So, we recreated the controlfile and decided to recover using backup controlfile option (until cancel or until change or until time). However, to my surprise, when I listed the checkpointchange# and checkpoint_time columns of v$datafile_header, I found the difference in timestamps ranging 24 hrs -- which menat we had to apply 24 hrs of archiving to bring all dbfs to a consistent state (before it allowed us to open the db) -- so far so good, it worked that way.
My question is that, while I trigger a lvl0 RMAN backup at say 4PM on 27th July and it finishes on 4PM on 28th July (as I am directly pushing around 1 TB data to tapes) -- what is the consistency timestamp of this backup. In other words, is this backup available for a restore of 4PM 27th July? Or is it available only for a restore of 4PM 28th July and later ...?
Isn't it that RMAN works at block level and when I fire a backup it should give me all the dbf's of that timestamp..?
Appreciate your inputs/comments.

When taking a online backup with RMAN, the datafiles backup is always inconsistent because it needs the archived redo logs generated during the backup to be consistent: http://download.oracle.com/docs/cd/B19306_01/backup.102/b14192/intro005.htm#sthref70.
A full online backup can only be used to restore the database to a point in time starting after backup end: if a backup starts at 4PM on 27th July and it finishes on 4PM on 28th July you can use it to restore the database to a point in time after 28-JUL 4PM.

RMAN : Backup to a tape.

Hi every Body,
I need please help on how to backup to a tape.
Exactly how to configure the MML interface ?

You can consider Oracle Secure Backup or a vendor from Oracle Backup Solutions Program.
Check Configuring RMAN to Make Backups to a Media Manager for configuration of RMAN and the vendor specific documentation for the media management configuration.

Rman backup failure, and is generating a large number of files.

I would appreciate some pointers on this if possible, as I'm a bit of an rman novice.
Our rman backup logs indicated a failure and in the directory where it puts its files there appeared a large number of files for the 18th, which was the date of the failure. Previous days backups generated 5 files of moderate size. When it failed it generated between 30 - 40 G of files ( it looks like one for each database file ).
The full backup is early monday morning, and the rest incremental :
I have placed the rman log, the script and a the full directory file listing here : http://www.tinshed.plus.com/rman/
Thanks in advance - George
-rw-r----- 1 oracle dba 1073750016 Jan 18 00:03 database_f734055071_s244_s1
-rw-r----- 1 oracle dba 1073750016 Jan 18 00:03 database_f734055096_s245_s1
-rw-r----- 1 oracle dba 1073750016 Jan 18 00:03 database_f734573008_s281_s1
-rw-r----- 1 oracle dba 1073750016 Jan 18 00:03 database_f734055045_s243_s1
-rw-r----- 1 oracle dba 524296192 Jan 18 00:03 database_f734055121_s246_s1
-rw-r----- 1 oracle dba 1073750016 Jan 18 00:03 database_f734055020_s242_s1
-rw-r----- 1 oracle dba 4294975488 Jan 18 00:02 database_f734054454_s233_s1
-rw-r----- 1 oracle dba 4294975488 Jan 18 00:02 database_f734054519_s234_s1
-rw-r----- 1 oracle dba 4294975488 Jan 18 00:02 database_f734054595_s235_s1
-rw-r----- 1 oracle dba 4294975488 Jan 18 00:02 database_f734054660_s236_s1
-rw-r----- 1 oracle dba 4294975488 Jan 18 00:02 database_f734054725_s237_s1
-rw-r----- 1 oracle dba 4294975488 Jan 18 00:02 database_f734054790_s238_s1
-rw-r----- 1 oracle dba 209723392 Jan 18 00:02 database_f734055136_s247_s1
-rw-r----- 1 oracle dba 73408512 Jan 18 00:02 database_f734055143_s248_s1
-rw-r----- 1 oracle dba 67117056 Jan 18 00:02 database_f734055146_s249_s1
-rw-r----- 1 oracle dba 4194312192 Jan 18 00:02 database_f734054855_s239_s1
-rw-r----- 1 oracle dba 2147491840 Jan 18 00:02 database_f734054975_s241_s1
-rw-r----- 1 oracle dba 3221233664 Jan 18 00:02 database_f734054920_s240_s1
drwxr-xr-x 2 oracle dba 4096 Jan 18 00:00 logs
-rw-r----- 1 oracle dba 18710528 Jan 17 00:15 controlfile_c-1911789030-20110117-00
-rw-r----- 1 oracle dba 1343488 Jan 17 00:15 database_f740621746_s624_s1
-rw-r----- 1 oracle dba 2958848 Jan 17 00:15 database_f740621745_s623_s1
-rw-r----- 1 oracle dba 6415990784 Jan 17 00:15 database_f740620829_s622_s1
-rw-r----- 1 oracle dba 172391424 Jan 17 00:00 database_f740620814_s621_s1

george3 wrote:
Ok, perhaps its my understanding of RMAN that is at fault. From the logs :
Starting recover at 18-JAN-11
channel m1: starting incremental datafile backup set restore
channel m1: specifying datafile copies to recover
recovering datafile copy file number=00001
name=/exlibris1/rmanbackup/database_f734055020_s242_s1
recovering datafile copy file number=00002
name=/exlibris1/rmanbackup/database_f734055045_s243_s1
it seems to make backup copies of the datafiles every night, so the creation of these large files is normal ?Above results indicate that you have full (incremental level 0) backup(all datafiles copies ) and there happen update/recover (applying incremental level 1) backup.So there was happen applying */exlibris1/rmanbackup/database_f734055045_s243_s1* inremental backup to full(level 1) backup.And size should be normal
Why is it making copies of the datafiles even on days of incrementals ?
Because after getting level 1 backup and need applying and every day will apply one incremental backup.

Rman backup on TEVOLI tape

Hi,
Can anybody help me.TSM tape configured for linux RAC oracle database for rman backups.once i start backup its giving me followng error.can anybody provide me help
Recovery Manager: Release 10.2.0.1.0 - Production on Fri Aug 18 20:49:36 2006
Copyright (c) 1982, 2005, Oracle. All rights reserved.
RMAN> connect target;
connected to target database: IPO (DBID=1999149714)
RMAN> BACKUP DATABASE PLUS ARCHIVELOG;
Starting backup at 18-AUG-06
current log archived
using target database control file instead of recovery catalog
RMAN-00571: ===========================================================
RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
RMAN-00571: ===========================================================
RMAN-03002: failure of backup plus archivelog command at 08/18/2006 20:50:19
ORA-19554: error allocating device, device type: SBT_TAPE, device name:
ORA-27000: skgfqsbi: failed to initialize storage subsystem (SBT) layer
Linux Error: 2534: Unknown system error
Additional information: 7011
ORA-19511: Error received from media manager layer, error text:
SBT error = 7011, errno = 2534, sbtopen: system error
RMAN>

Possible solution is:
The default path where TDPO looks at is '/usr/tivoli/tsm/client/api/bin64'
even if the file DSM.SYS exists in another directory.
The read permission of the file '/usr/tivoli/tsm/client/api/bin64/dsm.sys'
was incorrect:
-rw-r----- i.e. "OTHERS" which is Oracle isn't allowed to read the
file /usr/tivoli/tsm/client/api/bin64/dsm.sys
After changing permissions on /usr/tivoli/tsm/client/api/bin64/dsm.sys in order
to allow oracle to use it then the backup works fine.
$ chmod +r /usr/tivoli/tsm/client/api/bin64/dsm.sys
The problem is on the TSM side. If the solution does not work you have to check TSM logs for further error messages.
Werner

RMAN BACKUP ERROR in TAPE

Dear Friends,
I am taking a RMAN BACKUP of ORACLE 9i with AIX4.3 with TAPE DRIVE
When i execute the script
RMAN> RUN{
allocate channel c1 type 'SBT_TAPE';
backup tablespace users;
It gets struckup for more than two hours it didn't prompt any errors.
allocated channel: c1
channel c1: sid=13 devtype=SBT_TAPE
channel c1: MMS Version 2.2.0.1
Starting backup at 11-APR-02
channel c1: starting full datafile backupset
channel c1: specifying datafile(s) in backupset
input datafile fno=00007 name=/gloqm/oradata/ORA9011/users01.dbf
channel c1: starting piece 1 at 11-APR-02
LSM was intalled, Is it right to mention the tape is 'SBT_TAPE' or i will have to mention something different.
Could any one help
Thanks in Advance
gandhi

For RMAN topics, please use: {forum:id=74}.

RMAN Backup Direct To Tape

Hi,
Can anyone out there give me assistance with setting up my database in order to backup it up directly to tape, via RMAN. We have the following setup:
MS-Windows NT4.0 SP6.0a
Oracle 8.1.7
Veritas Backup Exec for Windows NT Ver7.3
Tape drive is a Compaq DLT 40/80 (x2) with a TL891 mini library.
Any help in this matter would be greatly appreciated.
Many thanks
Carl Scotson
DBA - West Of Scotland Water

If you don't already know the syntax for allocating a channel for a backup to tape here it is:
ALLOCATE CHANNEL <channel_name> type 'sbt_tape';
As for sending the backup to tape instead of the storage pool that will have to be setup on the media manager side. We do that for our backups with Tivoli, but it was setup to go directly to tape through the Tivoli media manager software.

Is RMAN backup directly to tape faster then to disk?

A coworker claims that performing a full backup to tape is twice as fast as opposed to disk. I cant find any documentation to support this claim. Any thoughts one way or another?

user10784896 wrote:
A coworker claims that performing a full backup to tape is twice as fast as opposed to disk. I cant find any documentation to support this claim. Any thoughts one way or another?Tell him to prove it. There are too many variables for someone to just make a claim like that without qualification. I'd say that - at best - he once clocked one paticular backup to tape as faster than one particular backup to disk.
That said, if the tape really were faster (even incrementally faster, not to mention "twice as fast") I'd say there was some serious mis-configuration in the disk backup - either in the specification or the layout of the disk system itself, because left to its on devices disk is inherently faster. Look at the published I/O specs of the disk vendor vs. the tape vendor.

Restoring RMAN backup from tape beyond retention policy ??

Hello Friends,
I just stuck in some basic finding. Need your help to resolve the questions.
How we can use RMAN backup to restore oracle database when the RMAN backup pushed to tape already and being a OLD backup ( before retention policy date).
Just to elaborate a more.. Say my retention policy is 3 days. I want to restore from a old backup like 30 days old.
So surely the requesting backup is behind retention date and has been pushed to tape.
Please suggest.
Thanks in advance.
Regards
Sinha.

dba.arnab wrote:
Hello Friends,
I just stuck in some basic finding. Need your help to resolve the questions.
How we can use RMAN backup to restore oracle database when the RMAN backup pushed to tape already and being a OLD backup ( before retention policy date).
Just to elaborate a more.. Say my retention policy is 3 days. I want to restore from a old backup like 30 days old.
So surely the requesting backup is behind retention date and has been pushed to tape.
Please suggest.
Thanks in advance.
Regards
Sinha.
Assuming the actual backup files still exist (the only thing actually deleted is the record of them in the catalog), you can make those files available and use the rman 'catalog' command to re-catalog them.

How to check the size of rman backups to tape

instead of asking tsm person to check, how do I check how big is a rman backup sent to tape?

query to find the backupset sizes
SQL> select ctime "Date"
2       , decode(backup_type, 'L', 'Archive Log', 'D', 'Full', 'Incremental') backup_type
3       , bsize "Size MB"
4 from (select trunc(bp.completion_time) ctime
5          , backup_type
6          , round(sum(bp.bytes/1024/1024),2) bsize
7     from v$backup_set bs, v$backup_piece bp
8     where bs.set_stamp = bp.set_stamp
9     and bs.set_count = bp.set_count
10     and bp.status = 'A'
11     group by trunc(bp.completion_time), backup_type)
12 order by 1, 2;
Date      BACKUP_TYPE    Size MB
03-JUL-10 Archive Log       7.31
03-JUL-10 Full             29.81
03-JUL-10 Incremental    2853.85
04-JUL-10 Archive Log       3.59
04-JUL-10 Full              7.45
04-JUL-10 Incremental       3.05

RMAN restore from monthly tape backup

hello!
i have one more question regarding rman and tape backups: we don't want to store the rman recovery information in a separate database, so all the information is in the control files. but for revision we need to save one full backup per month. in addition to the databases we need files from different servers etc. on this tape-backups. so we decided to keep one tape set per month in a safe place -> is it possible to restore a rman backup from this tapes, e.g. 3 years afterwards (on a different machine)? thanks.
bye,
christian

Hi christian,
Can you give more info?
As I understand it you do not want to use a recovery catalog ("we don't want to store the rman recovery information in a separate database").
But if you store the rman backup info in the controlfile only you will limit the amount of rman information to your init parameter control_file_record_keep_time.
Do you make rman backups to tape directly using a MML or do you make rman backups to disk which than get copied to tape (OS backup)?
Regards,
Tycho

RMAN backup slow

Hi,
I have a Production Database (10204) running on HP Unix (B.11.11).
Daily morning 6AM, a full RMAN backup runs. The db size is only around 225GB. But it takes 4.5 hours to complete.
The syntax which I use is given below.
RMAN> run {
     allocate channel ch1 type "sbt_tape";
     allocate channel ch2 type "sbt_tape";
     allocate channel ch3 type "sbt_tape";
send
'NB_ORA_CLIENT=jjcprd04-back.backup.ncsus.jnj.com,NB_ORA_POLICY=JJCPRD04-RMAN';
     backup
     format 'bk_%s_%p_%t'
     (database include current controlfile);
     sql 'alter system archive log current';
     backup
     format 'arch_%d_%s_%p_%t'
     (archivelog all delete input );
Is there a way to increase the speed of this backup? Currently the backup is way too slow as it takes 4.5 hours to copy 225GB of data!
Thanks!

I'll start by stating that backups directly to tape are generally quite slow.
I'd still recommend moving to disk backups though... and just archiving it to a tape...Hi Jony,
You can of course backup to disk. Thats a common solution. But you have to deal with space more than everything. In addition to that backing up to disk does not free you from moving/copying your data - depending on your backup strategy - to a long-term storage.
So you have two different and not conjunct backup operations: the rman backup itself and the job which backups the rman area to another media.
In recovery cases you might end up with RMAN requesting older backupsets which are not on disk anymore. This will most probably lead to recovery failures and manual intervention which is not desirable in recovery scenarios.
Another point is the control of the data flow. Backing up via a MML (i.e. "to tape") does not necessarily mean "to tape". Instead it means "to the backup software" which is responsible for managing the backup data. Common scenarios involve backing up the full backups via SAN directly to tape and backing up the archivelogs or incremental backups over LAN which will be stored by the backup software on a hard disk and later copied and/or moved on to tape. The data flow (to SAN / to disk / whereever) is solely controlled by the backup software.
The advantage here is one continuous backup and restore job. The backup software deals with "where is my saveset located" and requests automatically the needed tapes or disks in recovery scenarios. This makes recoveries far more flexible and comfortable.
Regarding the throughput i have set up a rather large OLTP database ( 12 TB currently) which is backed up with EMC Networker. Full backups are written via SAN directly to tape. The average speed with four LTO-3 drives is approx. 550 MB/s (the tape drives compress the data in hardware; the database itself is text-only and seems rather good compressable). I guess thats not too bad.
you can also compress the BACKUPSETS (this generates a little more CPU but less network IO)He said he is on HP-UX. These kind of CPUs tend to be not that fast. Using compression is an option worth testing but there is another point to test when doing compressed backups: The RESTORE TIME.
Ronny Egner
My Blog: http://blog.ronnyegner-consulting.de

Rman backups delete

What commands would I use to do the following:
Daily I take a full compressed backup. I currently use:
backup as compressed backupset database;
And backup my archivelogs. I currently use:
backup archivelog all;
Then I want to keep 2 days of backup and archivelogs on disk and maybe 7 days of backup and backed up archivelogs on tape. How do I accomplish this part?

For what you are doing I like to utilize a storage manager product with my tape system where the files on disk are brought to tape automatically and stub files left in place where the files appear to be in the directory but are really on tape. This way I can just manage RMAN backup pieces and execute dsik based backups normally but manage my disk space better, manage RMAN as normal because as for as RMAN is concerned the files are there on disk. Tivoli Storage Manager (TSM) has a product that does this called HSM, other vendors have Hierarchical Storage Management (HSM) products as well. They are usually pretty inexpensive and remove complication related to managing the backup pieces between disk and tape with RMAN. You get the best of both worlds without having to manage anything differently then just disk based backups and you can use less storage.

RMAN backup compression and tape hardware compression

Similar Messages

Maybe you are looking for