RECOVERY IN NOARCHIVE LOG MODE

Dear all,
Please help me in solving this recovery issue :
My database is : oracle 8i server ent.edition
database runs on windows 2000
db confgiured in no archive logmode
when I start the database today :
am getting the following error :
O/S-Error: (OS 23) Data error (cyclic redundancy check).'
error 1242 detected in background process
ORA-01242: data file suffered media failure: database in NOARCHIVELOG mode
ORA-01114: IO error writing block to file 2 (block # 1936)
ORA-01110: data file 2: 'c:\ORACLE\ORADATA\ORACLE\RBS01.DBF'
ORA-27072: skgfdisp: I/O error
OSD-04008: WriteFile() failure, unable to write to file
O/S-Error: (OS 23) Data error (cyclic redundancy check).
this tablespace rbs is defined only for rollback segments..
since db in noarhicve log mode, I cant able to take the datafile offline.. I tried doing
ALTER DATABASE RECOVER
which resulted in the above error .
I have valid 1 month old cold backup..
when I try to drop the tablespace, it fails with the above error.. .
Please guide me as this is the production database..
Thanks in advance
Please help me
Kai

Were you able to achieve a NORMAL or IMMEDIATE shutdown of the database ?
In that case the Rollback segments are not required at startup.
So,
1 comment out the entry "rollback_segments=...(namesofrollbacksegments)"
from the init.ora
2. connect / as sysdba and STARTUP MOUNT;
3. ALTER DATABASE DATAFILE '<filenameforRBS>' OFFLINE DROP ;
4. ALTER DATABASE OPEN;
Additional steps : CREATE ROLLBACK SEGMENT SYS_RBS01
TABLESPACE SYSTEM;
ALTER ROLLBACK SEGMENT SYS_RBS01 ONLINE;
5. DROP TABLESPACE RBS ; (check the name of the tablespace)
6. CREATE TABLESPACE RBS (or NEWRBS) DATAFILE ....
7. CREATE ROLLBACK SEGMENT ........ and create multiple segments in
the new RBS tablespace
8. SHUTDOWN
9. Edit init.ora to set the rollback segment names again
10. STARTUP
HOWEVER if your database shutdown was not normal or immediate but an
abort, Oracle would try to do an Instance Recovery. In that case, you would
need to use "_corrupted_rollback_segments" and may lose data consistency.
Check MetaLink for notes on that. But first check your shutdown and see if
you can follow the steps suggested by Sybrand and me.
Added additional steps for SYS_RBS01 in the SYSTEM
tablespace for any actions on the RBS tablespace.
Message was edited by:
Hemant K Chitale

Similar Messages

  • Why we cannot take hot backup if database is in noarchive log mode

    Hi,
    I am aware that if database is in noarchive log mode, we cannot take hot backups and only cold backup is possible.
    I would like to know the technical reason behind this restriction?
    Thank You
    Sarayu

    Hot backups are fuzzy backups, inconsistent, in other words, since something is always happening in the database.  When you recover, you restore data files and then apply redo to make the transactions consistent.  You can do a complete recovery or recover to a point in time.  So where does the redo come from?  That's what we call archiving redo logs.  When the online redo gets full, it gets archived.
    In the case of an instance crash, the redo is there in the online redo logs, so Oracle can recover automatically.  Anything beyond that, having to do with storage media, is a media recovery, and requires those archived logs.  So unless you have some other way to get your data back, always run in archivelog mode.
    It is really important to understand the concepts.  Please read the docs.
    http://docs.oracle.com/cd/E11882_01/server.112/e25789/cncptdba.htm#CNCPT031
    http://docs.oracle.com/cd/E11882_01/backup.112/e10642/rcmintro.htm#i1005488
    It may be worth your while to get a third party backup and recovery book too.

  • RMAN backup in noarchived log mode

    Hi,
    can we take RMAN backup in noarchived log mode?
    If yes, how? just run backup database command or ?
    thanks

    You can use RMAN to backup a database in noarchivelog mode but you need to shutdown and mount the database. e.g.:
    rman target /
    RMAN> shutdown immediate
    RMAN> startup mount
    RMAN> backup database

  • Recovery issue in noarchive log

    Hi,
    i need recovery for a database which is running in noarchive log mode.The dbas im my org take cold backup so they donot put the db in archive log mode. Now the database needs recovery it gets mounted but doesnot open and the db server asks me to apply the archive logfile which i donot have.i have also set the allowresetlogs_corrution to true. after restoring and performing the cancel based recovery the oracle server asks me for archive logfiles.i cannot perform incomplete recovery since the logifiles have been overwritten for a very long extend.
    so at this stage how can i recover the database.
    please provide me the essential steps.
    I was provided this issue to solve.kindly help me.......
    Thanks and Reagards
    Ilamparithi.A.

    An inconsistent backup of a NOARCHIVELOG database is useless... Worse than useless, actually, if it provides a false sense of security that the database has been backed up and the data is somewhat safe.
    I'm afraid the answer is most likely that the database is gone forever. If this is particularly valuable data, Oracle Consulting does have a tool (as do the ORA-600 folks in DUDE) that can probably recover most of the data from the data files. But these are not cheap utilities-- unless this is particularly valuable data, it may be more cost effective to toss the backups, find what data you can from other sources (old backups, old exports, other databases, etc) and chalk the cost of reconstituting the rest of the data up to the cost of hiring DBAs that didn't know what they were doing.
    Doug Burns has a discussion on the [http://oracledoug.com/serendipity/index.php?/archives/789-DUDE,-Wheres-My-Data.html|DUDE data unloader]
    I'd strongly suggest opening a Metalink ticket just to cover your own butt here. I expect them to tell you the same thing, but it's obviously a lot easier to go to management to tell them they've lost their data or that they need to write a check with lots of 0's to get it back because Metalink said so rather than because some guy on the internet said so.
    Justin

  • Archive log mode Turn off

    Hi all DBA guys, Have a Nice day all
    If i change archive log mode to no archive log mode, internally whats happen my database?.
    Is this old archive log backup use for future if again convert no archive to archive log mode?
    What's happen flashback if i change archive log mode to no archive log mode?
    Regards
    S.Azar

    azarmohds wrote:
    Thanx lot,
    why i asked this question means, rightnow you posted, this is feature of 10g i.e flashback database confiured when archive log mode is enabled...,so in this case
    What's happen flashback if i change archive log mode to no archive log mode?S.Azar,
    You should be clear about things that when you read notes from web, there can be a language mismatch. So the best is that search first over the oracle docs.
    Now about your doubt, its a terminology mismatch which is happening here. To enable Flashback Database feature which is more persistent copy of the undo data , stored seperately in a FLB file, you MUST have to have archivelog mode configured in your db. Without this, you can't enable the Flashback Database feature. And you can't switch to noarchive log mode as well if you have Flashback Database mode on.
    The other thing ( which I guess Anurag was mentioning ) is the normal Flashback feature which relies on Undo data stored in the undo tablespace. This solely is based on teh undo data and doesn't depend on the archive/nonarchive log mode of the database. But this is limited to that data and to that time period up till which you can store the undo data in your undo tablespace.
    HTH
    Aman....

  • After setting the db to archive log mode what to do?

    Hi All,
    We have a database which is being used as production database since one week and running in NoArchive Log mode. Now i wanted to change that to Archive Log mode because i want to take hot backup using RMAN daily/weekly. My doubt is..after changing the db to Archive Log mode, can i take backup using RMAN (i got a RMAN script to take a hot backup of db)? which will contain all the backup with control files and spfiles? or do i have to do anything before taking hotbackup.
    Thanks,
    Praveen.

    Yes you are fine. BTW, how many archive destinations did you configure.
    Usually it is more than 1 on two different Disks.
    TEST CASE:
    NOARCHIVELOG Database
    Change it ARCHIVELOG
    Take a hot backup and try to restore/recover it.
    STATUS=SUCCES
    Can someone explain this to me please:-
    Ensure DB in noarchivelog mode
    SQL> archive log list
    Database log mode No Archive Mode
    SQL> startup mount;
    SQL> alter database archivelog;
    SQL> alter database open;
    -- DID Not take a COLD BACKUP
    -- Right after switching the mode I am taking hot backup
    RMAN> backup incremental level=0 database ;
    RMAN> backup archivelog all delete all input;
    Nuke the datafiles
    RMAN> SET DBID=1299283748;
    RMAN> restore controlfile from autobackup;
    run
    set until logseq=79 ;
    restore database;
    recover database;
    sql 'alter database open resetlogs';
    Database is opened restored / recovered?
    I totally agree we should take a cold backup right after switching the mode. Can someone explain the above though.
    Message was edited by:
    gopalora

  • Recovery in noarchive mode

    My oracle database 11g running in no archive mode, Accidently some of the user datafiles move to some other location
    and them move back to same location with os command while database running in open mode , at that point in time redo log files changing frequently , Currently database is in open mode and can not be shutdown because of those files which have different version than other file and controlfile ,No backup is available , Is there any way to recover these files
    Regards
    obaid

    Hi,
    (1) As db is in no archive mode so you wont be able to apply redo and do recovery.since redo log file gets written in cyclic order after each switch log.
    (2) now replaced files are inconsistent with other files and recovery is not possible to make files consistent.
    i guess u r db corrupt. take help of oracle support OR
    Now one way u can offline replaced files using following command and keep running the database for other schema which are working file.
    alter database datafile 'location' offline ;
    Edited by: Rajesh.Rathod on Oct 3, 2012 5:20 AM
    Edited by: Rajesh.Rathod on Oct 3, 2012 5:38 AM

  • Different instructions for disable arch log mode on 11Gr2 RAC server?

    Hello all,
    I've run into a problem where I've lost my tape drive...and have no sysadmins to help.
    I don't want my RAC instances to run our of space and halt, so I'm planning to take them out of archive log mode, and just do exports daily till I can move them or get tape going again.
    This is easy enough with a non-clustered instance, but I'm reading around and finding conflicting information for doing it on a RAC system.
    In the Oracle® Real Application Clusters Administration and Deployment Guide
    11g Release 2 (11.2)...it states in simple terms:
    (http://download.oracle.com/docs/cd/E11882_01/rac.112/e16795/rman.htm#i474611)
    "In order for redo log files to be archived, the Oracle RAC database must be in ARCHIVELOG mode. You can run the ALTER DATABASE SQL statement to change the archiving mode in Oracle RAC, because the database is mounted by the local instance but not open in any instances. You do not need to modify parameter settings to run this statement."
    and thats about it.
    I've been researching and found a couple of other non-oracle official guides to this, where they describe a much more involved process, that seems to follow this path:
    1. sqlplus into one instance and change the cluster_database=false scope=spfile sid='specific_node_name';
    2. Shut down all instances, srvctl stop database -d <instance_name>
    3. Startup the instance you changed cluster_database on with sqlplus and startup mount;
    4 On this instance you do ALTER DATABASE NOARCHIVELOG;
    5. On same instance change the cluster parameter back: alter system set cluster_database=true scope=spfile sid='specific_node_name';
    6. Shut down this single instance
    7. Start all instances with srvctl start database -d <instance>
    I've found references to this at:
    http://oracle-dba-yi.blogspot.com/2010/12/enabledisable-archive-log-mode-in-11gr2.html
    and
    http://www.dba-oracle.com/bk_disable_archive_log_mode.htm
    Among other sites. I'm curious why the Oracle documentation on this doesn't mention all these steps?
    I'm guessing the longer version is the path I should take, but I wanted to ask here first if this is correct?
    I'm on Oracle 11Gr2....hasn't been patched with latest patchset, running on RHEL5, and is a 5 node cluster.
    Thank you in advance,
    cayenne
    Edited by: cayenne on Oct 21, 2011 11:51 AM

    Fiedi Z wrote:
    There are couple things you need to consider
    - export daily is not a backup strategy
    - you're risking your enterprise company by disabling archivelog
    your company have 5 RAC nodes I assume this is mid-to-large company, question you might ask to yourself Is your company really desperately don't have any available disk space for you to backup to temporary location / server?
    However if you still insist and persistent with your strategy then follow the links you have, that is how to disable archivelog in RAC
    CheersThank you everyone for the comments.
    This is a DEV environment...and they are planning to move this all to a new facility where we won't have the powerouttages and old defunct equipment.
    Right now...I do not have drive space to put all of this. I've informed them of the risks of not having point in time recovery. I really don't see any other choice on this....I don't want to run noarchive either...but I've been without tape to move the logs off for days now..and even with low traffic, I'm afraid they will fill and I'll have databases halting.
    I think at this point, and again, this is not production data....I'm going to have to go with daily exports and that will have to do me till I can get these servers 'moved' to a new facility soon.
    Again, thank you for the comments!!!
    cayenne

  • Oracle 10g ASM converting noarchive log to archive log

    DATABASE Details
    Oracle Database 10g Release 10.2.0.4.0 - 64bit Production
    PL/SQL Release 10.2.0.4.0 - Production
    CORE 10.2.0.4.0 Production
    TNS for 64-bit Windows: Version 10.2.0.4.0 - Production
    NLSRTL Version 10.2.0.4.0 - Production
    Mode
    Database log mode No Archive Mode
    Automatic archival Disabled
    Archive destination USE_DB_RECOVERY_FILE_DEST
    Oldest online log sequence 26269
    Current log sequence 26274
    SQL>
    TYPE :
    DW Batch Process
    Redo log groups
    6 groups with 2 files size each 2GB Aprx. 400gb redo log generates per day
    OS : WINDOWS 2008 SERVER
    Current Size
    SQL> select sum(bytes)/(1024*1024*1024) from v$datafile;
    SUM(BYTES)/(1024*1024*1024)
    1003.86945
    ASM Details
    SELECT GROUP_NUMBER,NAME FROM V$ASM_DISKGROUP;
    GROUP_NUMBER NAME
    1 KK_DATA
    SQL> SELECT dg.name AS diskgroup, SUBSTR(c.instance_name,1,12) AS instance,
    2 SUBSTR(c.db_name,1,12) AS dbname, SUBSTR(c.SOFTWARE_VERSION,1,12) AS software,
    3 SUBSTR(c.COMPATIBLE_VERSION,1,12) AS compatible
    4 FROM V$ASM_DISKGROUP dg, V$ASM_CLIENT c
    5 WHERE dg.group_number = c.group_number;
    DISKGROUP INSTANCE DBNAME SOFTWARE COMPATIBLE
    KK_DATA +asm         KK   10.2.0.4.0   10.2.0.0.0
    Currently DB in noarchive mode, Need to put in archive Mode
    What are additional precautions need to take in case of ASM for managing archive mode

    I would recommend creating a new ASM diskgroup and assign Flash recovery area to this group. This is part of ASM best practices as in case your Data diskgroup goes, you can use RMAN backups/archivelogs for recovery purpose.
    -Amit
    http://askdba.org/weblog/

  • Maxdb log mode

    hello experts,
          I need to change the log mode which is in "overwrite mode"
        currently. and increase LOG_IO_QUEUE size.
       how can i do it.
       our maxdb version is 7.6 and O/S linux 2.6.16
       plz suggest steps .
    thanks and regards

    Dear Kavitha,
    -> Please review the SAP notes::
                    869267        FAQ: MaxDB LOG area
         <  "35. Can the log area be overwritten cyclically without having to make a log backup?"
            "52. How large should the LOG_IO_QUEUE be when configured? " >
                 719652     Setting initial parameters for liveCache 7.5 or 7.6
                 819641     FAQ: MaxDB Performance
         If you are SAP customer, you are able to read the SAP notes.
    -> The MAXDB documentation also give you answers on the reported questions::
                http://maxdb.sap.com/documentation/ -> Open the SAP MaxDB 7.6 Library
                 ->  Glossary
              "Changing the Values of Database Parameters" at
                      http://maxdb.sap.com/doc/7_6/9b/e6dc41765b6024e10000000a1550b0/content.htm
              "Log Queue"
                    http://maxdb.sap.com/doc/7_6/23/c806c81e20f946a59a421e01c42c3b/content.htm
             < The new value of the LOG_IO_QUEUE parameter will be activated only after
                the database will be restarted from offline mode. >
              "Displaying and Changing Database Parameters" using DBMGUI tool at
                    http://maxdb.sap.com/doc/7_6/84/d8d198570411d4aa82006094b92fad/content.htm
    -> Please pay attention ::
          "Automatic overwrite of the log area without log backups is NOT    
               recommended for production operation. Since no backup history exists 
               for the following changes in the database, you cannot track transactions
               in the case of recovery. "    
          Run dbm command 'param_getexplain LOG_IO_QUEUE'.
              May be the log devspaces can be moved to a faster disk, to accelerate the physical log I/O.
    -> What is the version of the database? < Please also give the patch and build number >
         Why do you need to change the log mode which is in "overwrite mode"
          and increase the value of the database parameter LOG_IO_QUEUE?
    Thank you and best regards, Natalia Khlopina

  • When creating a tablespace why should we enable LOGGING when a database is already on ARCHIVE LOG mode

    Question :
    When creating a tablespace why should we enable LOGGING when a database is already on ARCHIVE LOG mode ?
    Example:
    Create Tablespace
    CREATE SMALLFILE TABLESPACE "TEST_DATA"
    LOGGING
    DATAFILE '+DG_TEST_DATA_01(DATAFILE)' SIZE 10G
    AUTOEXTEND ON NEXT  500K MAXSIZE 31000M
    EXTENT MANAGEMENT LOCAL
    SEGMENT SPACE MANAGEMENT AUTO;
    LOGGING: Generate redo logs for creation of tables, indexes and  partitions, and for subsequent inserts. Recoverable
    Are they not logged and not recoverable if we do not enable LOGGING? What is that ARCHIVELOG mode does?

    What is that ARCHIVELOG Mode Does?
    Whenever your database is in archive log mode , Oracle will backup the redo log files in the form of Archives so that we can recover the database to the consistent state in case of any failure.
    Archive logging is essential for production databases where the loss of a transaction might be fatal.
    Why Logging?
    Logging is safest method to ensure that all the changes made at the tablespace will be captured and available for recovery in the redo logs.
    It is just the level at which we defines:
    Force Logging at DB level
    Logging at Tablespace Level
    Logging at schema Level
    Before the existence of FORCE LOGGING, Oracle provided logging and nologging options. These two options have higher precedence at the schema object level than the tablespace level; therefore, it was possible to override the logging settings at the tablespace level with nologging setting at schema object level.

  • Putting a Oracle RAC in  to archive log mode??

    Our RAC is a 10gR2 Three(3) node RAC with a SAN as the shared storage. cluster is installed on linux. at the moment the RAC is in noarchieve log mode.
    SQL> archive log list;
    Database log mode No Archive Mode
    Automatic archival Disabled
    Archive destination /u01/app/oracle/product/10.2.
    0/db_1/dbs/arch
    Oldest online log sequence 665
    Current log sequence 666
    we created raw partitions in SAN and use ASM to manage shared storage.
    when running DBCA at "specify locations for the database files to be created" step we selected "use oracle managed files" and gave database area = +DATA/
    then at choose the recovery options for the database" step we ticked specify flash recovery area and gave Flash Recovery AREA = +BACKUP/ and flash recovery area = 10240 MB. we didnt click "enable archiving" option. hence the RAC DB is in NOARCHIEVELOG mode.
    i found some guide to put RAC in archive log mode which i have given below....
    ALTER SYSTEM SET log_archive_start=TRUE SCOPE=spfile;
    ALTER SYSTEM SET log_archive_dest_1='location=/u01/oradata/MYSID/archive/' SCOPE=spfile;
    ALTER SYSTEM SET log_archive_format='arch_%t_%s.arc' SCOPE=spfile;
    Since we need to mount the database in exclusive mode we must also alter the following parameter:
    ALTER SYSTEM SET cluster_database=FALSE SCOPE=spfile;
    From the command line we can stop the entire cluster using:
    srvctl stop database -d MYSID
    With the cluster down we can connect to a single node and issue the following commands:
    STARTUP MOUNT;
    ARCHIVE LOG START;
    ALTER DATABASE ARCHIVELOG;
    ALTER SYSTEM SET cluster_database=TRUE SCOPE=spfile;
    SHUTDOWN IMMEDIATE;
    Notice that the CLUSTER_DATABASE parameter has been reset to it's original value. Since the datafiles and spfile are shared between all instances this operation only has to be done from a single node.
    From the command line we can now start the cluster again using:
    srvctl start database -d MYSID
    my main question is about the setting of
    ALTER SYSTEM SET log_archive_dest_1='location= ????? ' SCOPE=spfile;
    since i have set +BACKUP/ as the FRA during installation how should i define the location of log_archive_dest_1???
    many thanks in advance.
    regards
    Dilan

    many thanks for the replies
    will the command you posted send the archive logs to the "+BACKUP/" location in my SAN.
    if yes how does oracle know that USE_DB_RECOVERY_FILE_DEST = BACKUP/ in my SAN. or how do i set oracle to use BACKUP/ as the USE_DB_RECOVERY_FILE_DEST??
    im a newbie to RAC so pls bear with me. also at the moment i get the following outputs from my database.... pls note the bold part below, is the archive destination pointing to a location in my local server?
    SQL> archive log list;
    Database log mode No Archive Mode
    Automatic archival Disabled
    Archive destination            /u01/app/oracle/product/10.2.0/db_1/dbs/arch
    Oldest online log sequence 665
    Current log sequence 666
    SQL> show parameter db_recovery;
    NAME TYPE VALUE
    db_recovery_file_dest string
    db_recovery_file_dest_size big integer 0
    SQL> show parameter archive
    NAME TYPE VALUE
    archive_lag_target integer 0
    log_archive_config string
    log_archive_dest string
    log_archive_dest_1 string
    log_archive_dest_10 string
    log_archive_dest_2 string
    log_archive_dest_3 string
    log_archive_dest_4 string
    log_archive_dest_5 string
    log_archive_dest_6 string
    log_archive_dest_7 string
    NAME TYPE VALUE
    log_archive_dest_8 string
    log_archive_dest_9 string
    log_archive_dest_state_1 string enable
    log_archive_dest_state_10 string enable
    log_archive_dest_state_2 string enable
    log_archive_dest_state_3 string enable
    log_archive_dest_state_4 string enable
    log_archive_dest_state_5 string enable
    log_archive_dest_state_6 string enable
    log_archive_dest_state_7 string enable
    log_archive_dest_state_8 string enable
    NAME TYPE VALUE
    log_archive_dest_state_9 string enable
    log_archive_duplex_dest string
    log_archive_format string %t_%s_%r.dbf
    log_archive_local_first boolean TRUE
    log_archive_max_processes integer 2
    log_archive_min_succeed_dest integer 1
    log_archive_start boolean FALSE
    log_archive_trace integer 0
    remote_archive_enable string true
    standby_archive_dest string ?/dbs/arch

  • Clone database with no archive log mode and downtime with new file structure

    We need to clone database with following business requirements :
    No downtime at all can be afford
    databases are in no archive log mode
    data-file locations need to be change to new mount points while cloning for space management factor
    Please suggest the best possible methods for same ?

    Can you post your version of oracle in 4 digits and OS for better understandig
    I dont think you can move the No Archive database to a new location without shutting down.
    You just want to move them to the new mount points is it?
    1. How big are your datafiles are?
    2.How do you backup this DB?
    When you move the datafilles then it becomes inconsistent and you need redo logs and probably archive logs to be there to make the file consistent or else it would be prompting for recovery. The chances are your redolog will be overwritten and hence you wont be able to recover.
    You need downtime.

  • Change Archive Log Mode, Restart instance, show ORA-12505 error!!!

    We are still use Oracle 9i.
    Now we want to use Symantec Backupexec 12 to backup oracle 9i.
    In accordance with Backupexec's requirements. must change recovery mode to Archive Log Mode.
    I use the Enterprise manager console to change it. and select restart immediately.
    a error message be display in the step 2 (starting up the database). see below
    ORA-12505: TNS:listener could not resolve SID given in connect descriptor
    I'm a new oracle user. Hope you give me some suggestion.
    THK.

    If database is down, then start it up directly login to server and
    sqlplus "/ as sysdba"
    startup;
    Read this thread for first:
    Re: ORA-12505, TNS:listener does not currently know of SID given in connect
    The main problem is Your listner doesn't know the sid in connect descriptor.
    Check
    lsnrctl services
    and look in tnsnames.ora file - is there such sercive defined in Your connection description.
    issue tnsping in order to check is it able to read and connect correctly tnsnames.ora entry.

  • Switching database to archive log mode

    1.When I switch the database to archive log mode,
    I have to do full offline(consistent) backup.
    2.When the database is working in archive log mode
    I can do inconsistent(online) backups.
    Why is there consistent backup needed after swiching
    to archive log mode, is there not inconsistent backup
    enough after switching (the databease operates
    in archive log mode and as in point 2 I can make inconsistent backups).

    Hello,
    Oracle recommands to do offline backup after switching to archivelog as an immediate crash-recovery procedure. Look at the following example:
    1/ You switch your base to archivelog.
    2/ Work is started right away.
    3/ Your DB crashes => You got your archived redo logs, but none of the other files.
    Now, let's put a
    1b/ Backup (offline) your database.
    Now, in 3, youg got everything you need to restore the base, because 3 can happen WHILE your are ONLINE Backing up the base.
    That's the point.
    Regards,
    Yoann.

Maybe you are looking for