Redo log and members groups

Similar Questions

• Redo Log and Supplemental Logging doubts related

  Hi friends,
  
  I am a student extra import logging in detail. Reading a lot of articles and the oracle documentation to this topic and redo logs. But couldnot find answers some doubts...
  Please help me to delete.
  
  Scenario: we have a table with primary key. And we execute a query update on the table that does not use the primary key column in a clause...
  Question: in this case, the roll forward records entry generated for changes made by put request to update contain primary column values... ?
  
  Question: If we have a table with primary key, we need to enable additional logging on primary column of this table? If so, in what circumstances, do we need to?
  
  Question: If we set up replication streams on this (having the primary key) table, why do we really need to enable saving of its complementary (I read documentation saying that flow requires some then more information., but in reality what information does it. Once again this question is closely related to the first question.)
  
  Please suggest also any good article/site that provide inside details of the redo log and additional logging, if you know.
  
  Kind regards
  Lifexisxnotxsoxbeautiful...

  (1) assuming that you do not update the primary key column and additional logging is not enabled, Oracle doesn't have to log in the primary key column in the log to roll forward, just the ROWID.

  (2) is rather difficult to answer without being tautological. You need to enable additional logging if and only if you have some use downstream for additional columns in the redo logs. Rivers and streams, these technologies built above are the most common reason to enable additional logging.

  (3) If you run an update as

  UPDATE some_table
    SET some_column = new_value
   WHERE primary_key = some_key_value
     AND <>
  

  and look at an update statement that LogMiner relies on newspapers in recovery in the absence of additional record, basically it would be something like

  UPDATE some_table
    SET some_column = new_value
   WHERE rowid = rowid_of_the_row_you_updated
  

  Oracle has no need to replay the exact SQL statement you issued, (and so he doesn't have to write the SQL statement in the redo log, he doesn't have to worry if the UPDATE takes a long time to run (otherwise, it would take so much time to apply an archived log as it did to generate the log) ((, which would be disastrous in a recovery situation) etc.). He just needs to rebuild the SQL statement of the information contained in the restoration by progression, which is just the ROWID and the columns that have changed.

  If you try to execute this statement on a different database (via the stream, for example) may the ROWID on totally different destination database (since a ROWID is just a physical address of a line on disc). So adding additional record tells Oracle to connect to the primary again key column and allows LogMiner / flow / etc. to rebuild the statement using the values of the primary key for the changed lines, which would be the same on the source database and destination.

  Justin

• Recreate the Member lost/damaged redo log in a group

  DB version: 11.2.0.4

  OS: Oracle Linux 6.5

  In this particular PB we have multiplexes recovery logs (2 members from each group of redo).

  If one member of a group of Redo log is corrupted then, regardless of whether the Member belonged to the Group that is CURRENT, ACTIVE or INACTIVE, can we just let GO and CREATE the file journal lost this group member by following the steps below?

  -Find the corrupted Member

  SELECT GROUP #, STATUS, MEMBER OF V$ LOGFILE WHERE STATUS = 'INVALID ';

  1 back UP the DB

  2. CHANGE THE SYSTEM CHECKPOINT;

  3 ALTER DATABASE CLEAR LOGFILE GROUP 3;  - , This step should not be executed. Right?

  4. ALTER DATABASE DROP LOGFILE MEMBER ' / oracle/oradata/wms/redo02.log';

  (Apparently, this removes not the logfile at the OS level, only the control file is updated in the DB)

  5 ALTER DATABASE ADD LOGFILE MEMBER ' / oracle/oradata/wms/redo02b.log' to GROUP 3.

  When I googled for steps, I found the table below in the official documentation of the Oracle. This table is applicable only if the ENTIRE REDO LOG disappeared.

  http://docs.Oracle.com/CD/E25054_01/backup.1111/e10642/osadvsce.htm#CACFJDBG

  I want just the steps if only 1 member of a multiplex group is lost/damaged.

  3. ALTER DATABASE CLEAR LOGFILE GROUP 3;  -This step should not be executed. Right?

  Yes. It's true. You can refer to the section 'Recovery after losing a member of the a multiplex online Redo Log group' under recovery after the loss of online redo log files: scenarios

  HTH

  -Jonathan Rolland

• Standby Redo logs and the Directory Structure in the Backup Site

  Hi guru

  I just want to confirm, I know that if the directory structure is different, I need to talk about these 2 settings in the file pfile

  on the main site:

  DB_CONVERT_DATAFILE = 'sleep', 'primary '.

  LOG_CONVERT_DATAFILE = 'sleep', 'primary '.

  On the secondary Site:

  DB_CONVERT_DATAFILE = 'primary', 'sleep '.

  LOG_CONVERT_DATAFILE = 'primary', 'sleep '.

  But I want to confirm this weather I have to deliver the full path of the directory in the two paramtere above:

  as:

  DB_CONVERT_DATAFILE = ' / u01/oracle/app/oracle/oradata/sleep ', ' / u01/oracle/app/oracle/oradata/primary.

  LOG_CONVERT_DATAFILE = ' / u01/oracle/app/oracle/oradata/sleep ', ' / u01/oracle/app/oracle/oradata/primary.

  Second Confusion: -.

  After that transfer standby Redo logs creates primary and made to sleep on the foregoing, mentioned the directory structure and after restoring the alongwith primary db backup, that ensures the control file will not affect the journal of physics again placed watch on the above mentioned location.

  Thanks in advance for your help

  vk82 wrote:

  In fact, I create the day before by using the RMAN Duplicate command. but where I am confused is the point if I transfer the backup that i taken the pri on the C:\backup_files path. and after that, I transfer the backups to the waiting on C:\backup_files. After that when I restore it will create datafile and other stuff in another directory I mentioned using DB_FILE_NAME_CONVERT as LOG_FILE_NAME_CONVERT. I think Yes but need your advice during the same period.

  Hello

  Yes, files should be created under the directory mentioned in the path of the parameters 'db_file_name_convert' and 'log_file_name_convert.

  Kind regards
  Shivananda

• Redo logs and crash of the disk where they are stored

  What happens when the disk where the redo logs are written becomes HS (e.g., a disk crash). I read somewhere that the database be shutdown. Is this right? You can ask to copy the recovery logs at 2 locations in order to improve the security of the system?
  
  Thanks in advance,
  Alexandre Bailly

  user8716187 wrote:
  What happens when the disk where the redo logs are written becomes HS (e.g., a disk crash). I read somewhere that the database be shutdown. Is this right? You can ask to copy the recovery logs at 2 locations in order to improve the security of the system?

  Thanks in advance,
  Alexandre Bailly

  Online redo logs can and should be multiplexed, with copies on separate physical devices. The details are in the ALTER DATABASE command, found in the SQL reference Guide. Additional information can be found by going to tahiti.oracle.com, drilling until your product (without name) and version, the by using the "search" feature to find something on "restore". A lot of information in the Administrators Guide.

  In addition, default Oracle will appoint the «redo_*.log» redologs ".Log" is an open invitation to ITS s to open the with a text editor or delete. After all, 'it's just the log file I named mine by the older convention of the "redo_*.rdo" just to reduce this kind of human error.

• Multiplex redo logs and slow down the e/s

  I sometimes use a San that gets saturated by a another client that I have no control over. My recovery on is newspapers the SAN and it causes all my customers to save waiting for synchronization of log file when it arrives. My overall IO is low, but customers commit frequently. In this scenario, I have no control over the San or other customer behaviours.
  
  What I'm considering is multiplexing by newspapers of recovery more fast local disk, but leaving a member of newspaper on the SAN because these discs are less likely to fail purely and simply.
  
  My question is how Oracle will behave during the SAN slows down - it will deal with the posting as 'fact' when the first newspaper finishes member in writing, or the last? I know that if writing an error then just Oracle will mark this member as dead, however in this case I get no errors, some delays.
  
  Edit: I have no control over the San or its configuration as well, I just have to work around the problem, while SAN team working on it.
  
  Published by: jhmartin1 on April 7, 2009 15:55

  Write completion wil be when more destinations slow two members signals, that Scripture is complete.

  Other thoughts.

  For a San, I expect to be when writing reaches a battery backed cache. So think of it as no adaptation write cache can be useful.
  You can increase the cache to write to your San?

  Obviously writing half as much for the san that can help.

  Make sure you use the SAME guidelines [http://www.oracle.com/technology/deploy/availability/pdf/OOW2000_same_ppt.pdf]

  Check if the problem is caused when backups are.

  Check a second time SAN and SUN firmware is up to date and ensure that best practices for san are respected etc.

  -I think that there is a problem for sun solaris I try to drag in the back of my mind by which a series of small writings are not well on the most massive. But Ithink that is direct attached tables and there is a workaround.

  HOEP this helps, bigdelboy.

• One group of standby Redo Log is ACTIVE

  Hi guys,.

  I have successfully configured a custodian of data between a primary database (oradb) and a database ensures Physics (oradb_s8).

  However, I have noticed that in V$ STANDBY_LOG only one group of standby Redo Log is ACTIVE, regardless of how many times I go log in the primary database.

  ' The following was stated in the documentation:

  When a switch of newspaper occurs on the database of the source again, redo incoming is then written to the next group of waiting for redo log, and the group used newspaper before again Eve is archived by a foreground ARCn process.

  Source.

  So, I guess that group of standby Redo Log is turned on when the Redo Log is enabled in the primary database.

  Could you please clarify it for me?

  It's the Oracle 11 g R2 (11.2.0.1) on Red Hat Server 5.2.

  On autonomy in standby:

  SQL > SELECT GROUP #, THREAD #, SEQUENCE #, ARCHIVED, STATUS FROM V$ STANDBY_LOG

  GROUP # THREAD # SEQUENCE # ARC STATUS

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

  4 1 248 YES ACTIVE <-this is the only group that is still ACTIVE

  5 1 0 NOT ASSIGNED NO.

  6 0 YES 0 UNASSIGNED

  7 0 YES 0 UNASSIGNED

  SQL > SELECT SEQUENCE #, APPLIED, FIRST_TIME NEXT_TIME

  V $ ARCHIVED_LOG

  SEQUENCE ORDER #;

  SEQUENCE # FIRST_TIM NEXT_TIME APPLIED

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

  232 06-06-SEPT.-15-15-SEP YES

  233 06-06-SEPT.-15-15-SEP YES

  234 06-06-SEPT.-15-15-SEP YES

  235 06-06-SEPT.-15-15-SEP YES

  236 06-06-SEPT.-15-15-SEP YES

  237 06-06-SEPT.-15-15-SEP YES

  238 06-06-SEPT.-15-15-SEP YES

  239 07-06-SEPT.-15-15-SEP YES

  240 YES 15-SEP-07 07-SEVEN.-15

  241 YES 15-SEP-07 07-SEVEN.-15

  242 YES 15-SEP-07 07-SEVEN.-15

  YES 243 15-SEP-07 07-SEVEN.-15

  244 YES 15-SEP-07 07-SEVEN.-15

  245 YES 15-SEP-07 07-SEVEN.-15

  246 YES 15-SEP-08 15-SEP-07

  247 15 - SEP - 08 IN-MEMORY 15-SEP-08

  16 selected lines.

  On the primary:

  SQL > SELECT SEQUENCE #, APPLIED, FIRST_TIME NEXT_TIME

  V $ ARCHIVED_LOG

  WHERE NAME = 'ORADB_S8 '.

  SEQUENCE ORDER #;

  SEQUENCE # FIRST_TIM NEXT_TIME APPLIED

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

  232 06-06-SEPT.-15-15-SEP YES

  233 06-06-SEPT.-15-15-SEP YES

  234 06-06-SEPT.-15-15-SEP YES

  235 06-06-SEPT.-15-15-SEP YES

  236 06-06-SEPT.-15-15-SEP YES

  237 06-06-SEPT.-15-15-SEP YES

  238 06-06-SEPT.-15-15-SEP YES

  239 07-06-SEPT.-15-15-SEP YES

  240 YES 15-SEP-07 07-SEVEN.-15

  241 YES 15-SEP-07 07-SEVEN.-15

  YES 07-SEP-15 242 07-SEP-15

  YES 243 15-SEP-07 07-SEVEN.-15

  244 YES 15-SEP-07 07-SEVEN.-15

  245 YES 15-SEP-07 07-SEVEN.-15

  246 YES 15-SEP-08 15-SEP-07

  247 NO. 15-SEP-08 08 - SEPT.-15

  If you have a large DML activity on primary, you see more than one group # in v$ active standby_log.

  RFS will always try to assign the next available standby log, because your group changes # 4 are already applied, it allocates this group again after the switch.

  Check metalink doc: bug 2722195 and 219344.1

• Controlfile and Redo Log on a disk group

  Hello!
  
  Controlfile and Redo Log on a disk group - it's good or bad?
  
  You have the best practices or Doc ID on Metalink?

  918027 wrote:
  You must have two control files on two separate disks and at least 2 members of each group redolog, each on a separate drive.

  http://docs.Oracle.com/CD/B10500_01/server.920/a96521/control.htm#4578

  No, you do not have "must have".

  You have not yet 'must have' more than a control queue.
  You do not have 'must have' more than 2 redo log groups, and these groups don't 'must have', but one file each.

  This is the "must have".

  Now, it's something else.

  You "should" have a minimum of 2 files of control, and that they "should" (not "needs") be on separate disks.
  You "should" have at least 2 files of members in each group of redo log, and that they "should" (not "needs") be on separate disks.

  And FWIW, I see no reason to keep necessarily completely separate redo file control file. They key is to have versions multiplexed of a particular type, separated from their 'twin '.

• relationship between redo log buffer, journal of redo and undo tablespace files

  What is the relationship between the redo log buffer, redo log files and undo tablespace?
  
  what I understand is
  
  redo log buffer is the logical area where all the news of recovery were stored until they are transferred by LGWR bank roll forward log online... but y at - it report any with the undo tablespace with these two?
  
  Please correct me if I'm wrong
  
  Thanks in advance

  redo log buffer is the logical area where all the news of recovery were stored until they are transferred by LGWR bank roll forward log online... but y at - it report any with cancellations

  tablespace with these two?

  There is a link between files redo log and buffer, but the undo tablespace is something else entirely.

  spend it here with this links

  REDO LOG FILES
  http://www.DBA-Oracle.com/concepts/redo_log_files.htm

  BUFFER REDOLOG
  http://www.DBA-Oracle.com/concepts/redo_log_buffer_concepts.htm

  UNDO TABLESPACE
  Undo tablespace to cancel files to undo or roll back uncommitted changes pray the database.

  Hope you understood.

• Redo Logs

  Running Oracle 11 g r2.

  I just want to confirm if Oracle recommends at least 3 redo logs to run the database?

  If so, how can I create the third (and fourth) redo logs (I believe the default is 2 recovery logs)?

  Thank you.

  Scott

  Running Oracle 11 g r2.

  I just want to confirm if Oracle recommends at least 3 redo logs to run the database?

  Post a link to the source of this information.

  If so, how can I create the third (and fourth) redo logs (I believe the default is 2 recovery logs)?

  You CAN'T do that. Do NOT modify Oracle based on your personal beliefs. You must use FACTS.

  What you believe has NOTHING to do with the reality of what can be your real system. You can only change your system.

  So first, you must determine how many newspaper groups and members your system log has already.

  The detail on the REDO log and groups members that exist on your system. The Oracle documentation at the address provided already has examples on how to do it.

• Question about how Oracle manages the Redo logs

  Hello
  
  Assuming a configuration which consists of 2 redo log groups (groups A and B), each group of 2 disks (disks A1 and A2 for Group A) and B1 and B2 disks for Group B. Additionally, assume that each redo log file resides by itself in a disk storage device and that the device is dedicated to her. So in the situation described above, there are 4 discs, one for each redo log file, and each disc contains nothing other than a redo log file. Also, assume that the database is in ARCHIVELOG mode and the files from archive is stored on another different set of devices.
  
  kind of graphically:
  

      GROUP A             GROUP B
  
        A1                  B1
        A2                  B2

  The question is: when the disks that make up the Group A are filled and Oracle switches to the disks in the Group B, can the group drives to take offline, perhaps even physically removed from the system if necessary without affecting the functioning of the database? Can the archiver process temporarily delayed until the disks (which have been removed) are presented online or is the DBA have to wait until the end of the process to archive a copy of the redo log file creating in archive?
  
  Thank you for your help,
  
  John.

  Hello
  A journal of the groups fall

  To remove a group of online redo logs, you must have the ALTER DATABASE system privilege. Before you delete a line redo log group, consider the following precautions and restrictions:

  * An instance requires at least two groups of files logging online, regardless of the number of members in the groups. (A group is one or more members.)
  * You can delete a group of newspapers online redo only if it is inactive. If you need to drop the current group, first force a log switch occurs.
  * Make sure a group of online redo logs is archived (if archiving is enabled) before dropping. To see if this happens, use the view LOG V$.

  SELECT GROUP #, ARCHIVED, STATUS FROM V$ LOG;

  GROUP # ARC STATUS
  --------- --- ----------------
  1 ACTIVE YES
  2. NO CURRENT
  3 INACTIVE YES
  4 INACTIVE YES

  Delete a group of newspapers online redo with the SQL ALTER DATABASE statement with the DROP LOGFILE clause.

  The following statement drops redo log group number 3:

  ALTER DATABASE, DROP LOGFILE GROUP 3;

  When a group of online redo logs is deleted from the database, and you do not use Oracle managed files, operating system files are not removed from the disk. Instead, control of the associated database files are updated to remove members of the Group of the database structure. After deleting a group of online redo logs, make sure the drop completed successfully and then use the command of operating system appropriate to delete the dropped online redo log files.

  When you work with files managed by Oracle, the cleaning of operating system files is done automatically for you.
  Your database will not be affected as you can work with 2 files, redo log in each group, as the minimum number of redo log in a database file is two because the process LGWR (newspaper writer) writes in the redo log in a circular way. If the process crashes because you have 2 groups only if you want to remove 1 Add a third and make that the current group and remove the one you want to be offline.

  Please refer to:
  http://download.Oracle.com/docs/CD/B10500_01/server.920/a96521/onlineredo.htm#7438
  Kind regards
  Mohamed
  Oracle DBA

• number of redo log files

  Hello
  What is the best practice to determine the number of redo log files,
  3 groups with 1 member of each or,
  2 groups with 2 members of each?
  which is optimal and recommended?
  Thank you very much

  Agree with Deepak.

  http://download.Oracle.com/docs/CD/B19306_01/server.102/b14231/onlineredo.htm#i1306224

  Read the link. It will help you.

  Concerning
  Asif Kabir

• Redo log size of the defect of EBS

  Hi all

  EBS R12.2.4

  11 GR 2

  Rhel6.5

  I noticed that the installation by dΘfaut of EBS has only two 1 GB size redo log groups.

  Is this acceptable to perform better?

  Your comment is very much appreciated.

  Thank you very much

  JC

  Protection for the primary database in three ways:

  Maximum availability: Transactions on the primary do not commit to redo information was written for the online redo log and newspapers awaiting restoration by progression of at least a place to sleep. If no location intelligence is available, it in the same way as the maximum performance mode until a standby is available again.
  Maximum performance: Transactions on the validation of primary as soon as redo information has been written for the online redo log. Transfer of information to roll forward to the standby server is asynchronous, so it has no impact on the performance of the primary.
  Maximum protection: the Transactions on the primary do not commit to redo information was written for the online redo log and newspapers awaiting restoration by progression of at least a place to sleep. If the standby not suitable location is available, the primary database stops.

  By default, a newly created Eve, the primary database is in maximum performance mode.

  I suggest to use the maximum performance with delay mode in the application of 15 minutes. You can drop the idea of delay if you are not comfortable with data loss

  concerning

  Pravin

• Standby redo logs

  Hi all

  For the installation of dataguard, supposed to docs create standby redo logs on the standby server.

  It does not mention to create logs to redo waiting on the primary server.

  So what happens if I pass of the day before?

  Then my primary db now becomes the day before db. But there is no waiting for redo logs?

  Kindly shed lights...

  Thank you very much

  JC

  Jenna_C wrote:

  Hi Sysas and ALL,

  the rule is to create a group of newspapers waiting to roll forward pending Eve more than the number of redo log group in the primary.

  Do you mean > the rule is to create a group of newspapers waiting to roll forward pending Eve more than the number of the Group of online redo logs in the primary?

  So, if I have 3 ONLINE redo log groups and 4 STANDBY redo log at the primary level, 3o + 4 s = 7

  I'll create 3 groups of online redo logs and 1 redo log group ENSURES the standby?     3O + 1 s = 4

  Why not just create everything as even with the main number?  3O + 4 s = 7 for sleep?

  Thank you.

  Sorry... If you have 3 groups of newspapers online redo (in primary and standby), and then create 4 STANDBY redo record groups (primary and standby).

  Concerning

• Create the REDO Logs after restoration of RMAN

  Version 11.2
  Platform: Solaris 10
  
  We restore the cold RMAN backup to a new server. It's a basic restore RMAN where DB name remains the same.
  
  I understand that RMAN backup online redo logs, and for good reason.
  
  We have restored our datafiles successfully. The control file still has the old entries so the query below shows old information about recovery logs
  
  -Display of the configuration of the log file the old control file data

  SYS > select * from v$logfile;         ------------> The below output shows the result of the source server
  
      GROUP# STATUS  TYPE    MEMBER                              IS_
  ---------- ------- ------- ----------------------------------- ---
           1         ONLINE  /DATAM/DB/oradata/slmps/redo1.log      NO
           2         ONLINE  /DATAM/DB/oradata/slmps/redo2.log      NO
           3         ONLINE  /DATAM/DB/oradata/slmps/redo3.log      NO

  I have to use ALTER DATABASE RENAME FILE command as below to update the control file.

  alter database rename file '/DATAM/DB/oradata/slmps/redo1.log' to '/data/oradata/slmps/slmps_log01.dbf' ;
  alter database rename file '/DATAM/DB/oradata/slmps/redo2.log' to '/data/oradata/slmps/slmps_log02.dbf' ;
  alter database rename file '/DATAM/DB/oradata/slmps/redo3.log' to '/data/oradata/slmps/slmps_log03.dbf' ;  

  Currently, in the new server, at the level of the BONE there is no physical file above to
  

  /data/oradata/slmps

  I need to create physically redo new logs using the command or the command ALTER DATABASE RENAME above key create a physical file at the level of the BONE?

  Salvation in spiral,

  Surely, RMAN will search for redo logs when RECOVER is issued.

  When you issue the command RECOVER after restore, RMAN does not check online redo logs. During RECOVERY, RMAN will try to apply the archivelogs or incremental L1 if it has been saved on the source server. RMAN Backup restorations online redo logs or never.

  Here OP tries to restore the backup from the cold to a new server. In this case the RECOVER command is not necessary. The example of the Oracle doc you mentioned should do a restore, recovery of a backup hot. In this case, during the RESTORATION, RMAN will try to apply archivelogs or L1 of the incremental backups.

  In both cases (cold backup restore or restore, recover from Hot Backup), it is the

  alter database open resetlogs
  

  command that will create the online redo log, so it is absent in the location specified by the control file. Because it is a restoration to a new server, recovery logs is missing because it has not been restored (because it has not been saved)