Redolog Buffer

Hi,
i know it is a silly question... sorry for asking silly question.
Redo is getting written to Redo Log Buffer and then written to Redo Log Files.
Redo Means, all changes happened to Data Blocks. Is it means, Information about the Changed data (Meta Data), is getting into Redo Log Buffer, which is used to reconstruct the data in crash recovery or actual changed data is getting written into Redo Log Buffer.
Please let me know, what is happening when data is changed in blocks. complete cycle interms of Buffer cache and redo log buffer...
Thank you very much in advance.....
Edited by: oraDBA2 on Jul 10, 2010 9:01 AM

Please let me know, what is happening when data is changed in blocks. complete cycle interms of Buffer cache and redo log buffer...I guess, if you have read the link given by Jonathan, it should be clear.Still, this ( in simple terms) happens,
1) The block is loaded into the buffer cache.
2)The transaction table within the Undo segment is updaetd with the transaction information. Also, the Undo block is loaded with the old information.
3)With the help of the concept, write ahead logging , the corresponding change vectors for the statement are loaded into the log buffer.
3)The statement is executed finally and the image in the buffer cache is changed according to the statment, for a delete, a delete of the row happens and for the update, an update happens.
4) The scn in in the transaction header of the block is updated as the transaction SCN and the transaction status is marked as active.
HTH
Aman....

Similar Messages

Disk IO for redolog buffer- isnt it a performance issue also ?

hi guys,
i am a noob but i have a curious question to ask.
hope you guys dont mind helping me out abit,
as we all know, disk io is an expensive operation, so oracle would rather keep the blocks in the db buffer cache as long as possible then to write them down everytime a commit occurs.
however
wouldnt writing and fetching from the redo log file into the redo log buffer an expensive operation as well ? especially it is a on 3 second basis ?
let say i just retrieve a empty block from my redo log file, fill in just 1 row. then it get written back to file. then again i make a update, the same block is retrieve again to fill up aganother row, then 3 second it is written back. wouldn't this be a expensive operation too ?
i know it must be done, to make sure every recorded changes make to the data blocks is written to disk as soon as possible as if we keep it long in the redo buffer, and the instance crash, we would have lost both the data and the changes recorded. so thats the only way out..
but is my theory right ?
thanks.
-noob

oraclewannabe,
redolog io is always completely sequential, and records are always appended.
datafile I/o is random or sequential.
as long as you don't locate your online redolog on i/o bound disks or raid-5 disks,
there shouldn't be a problem.
Sybrand Bakker
Senior Oracle DBA

Need help on recovery

Hello All,
We have a datagurad implmentated system. Yesterday due to some reasons Primary server crashed( it will take 1 or 2 day get back the server with filesystem) .
Now we have secondary system at diffrent location which is still up and running.
I want to understand following
Now secondary location is having last archive log file applied. but my Primary DB is having 200MB of redolog buffer.
1)what will happen if I will start the seconday DB? Is 200MB of data will be lost ?
2) Apart from log switch how archive log file will get generated ?
3) what is best solution to recover data up maximum level ?
Thanks & Regards,
Ajay Bhivsankar.

Ajay,
(A) What dataguard mode are you using?
(B) What log_archive_dest_n options did you specify on the primary?
(C) Are you using standby redologs on the standby?
(D) What is your version and platform?
If for (B) you were using LGWR, and for (C) you are using standby redologs, you probably have all the redo you can get in files on the standby. If for (B) you were using ARCH, then you are probably missing the most recent redo entries since the last log switch. If that is the case and you can't get the most recent redolog from the primary to copy, then you have lost data.
The log buffer is not particularly relevant to your concern, since it is almost always nearly empty. See the documentation below for the reasons:
[http://download.oracle.com/docs/cd/B19306_01/server.102/b14220/process.htm#sthref1530|http://download.oracle.com/docs/cd/B19306_01/server.102/b14220/process.htm#sthref1530]
Regards,
Jeremiah Wilton
Blue Gecko, Inc.
[http://www.bluegecko.net|http://www.bluegecko.net]

Process dbwriter and checkpoint

Hi,
It is known that when the redo log buffer is full at the third(1/3) , the datas which are in the redolog buffer are sent to the database's redo log files .
What happens then with the dbwriter process and the checkpoint process?Is there a new sequence number( SCN=system change number) created when the redo log buffer is sent to the database ?
Secondly, concerning the log_check_point_interval, can you explain to me what means that "in the number of redo log file blocks that can exist between an incremental checkpoint and the last block written to the redo log. This number refers to physical operating system blocks, not database blocks."
LOG_CHECKPOINT_INTERVAL specifies the frequency of checkpoints in terms of the number of redo log file blocks that can exist between an incremental checkpoint and the last block written to the redo log. This number refers to physical operating system blocks, not database blocks.
I don't exactly understand when occurs a log_checkpoint and what means that it concerns operating system blocks and not database blocks.
Does it mean it concerns the redo log buffer?
Thanks a lot for your answer.
Best regards.
Nathalie

DBWR does not write dirty buffers to the corresponding disk blocks until the Redo for those blocks (i.e. the changes made to the blocks) has been written by LGWR to the online redo log files. Therefore, Redo is always written to disk before modified blocks (whether they be table or index or undo blocks) are rewritten to disk.
The block size for online redo log files is NOT the database block size (which is specified by the instance parameter db_block_size). Redo log files are written direct to the OS and use the OS's block size -- which is typically 512bytes.
Thus, specifying LOG_CHECKPOINT_INTERVAL of, say, 100000 is an instruction to Oracle that "when the current online redo log file is about 50MB (100000 512byte blocks is 48.82125MB) written, issue a Checkpoint to DBWR so that DBWR also writes dirty buffers to disk". That way, every checkpoint interval is actually smaller than the actual online redo log file (a checkpoint will still be issued when the online redo log file is full and a switch logfile occcurs).
Hemant K Chitale

Problem in migration 7.3.4 - 8.0.6

I tried to migrate from Oracle 7.3.4.0.0 to Oracle 8.0.6.0.0 through Oracle Migration Assistance.
Everything was fine but after migration step when it was trying to start the database the log was showing ora-07304 illegal redolog buffer also found that old controlfiles were backedup to .old file. System could not generate a new control file. hence in starting the database it could not find the controlfile.
Did anybody face such problem?
Pl reply.
Regards,
Atul Jawalkar
null

Hi,
I've found the solution. I have to stop and start my listener each morning (i shutdown and save my dbs each night). now, it work fine !!
<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Originally posted by jean-guy avelin ([email protected]):
Hi all !
It seems I can't acces views from a jsp using jdbc :
when I try stmt.executeQuery("select * from dual"), it works. when i create a basic view on my dbserver ( create view toto as (select * from sys.dual) ) and I try this jdbc query stmt.executeQuery("select * from toto"), I get this error : ORA-12154
I use Oracle V7.3.4.5.0 on tru64Unix 5.0A on my oracle server. I've tried V8.1.7 and V3 jdbc drivers. Last year, I think my program (using views) ran fine with an oldest version of oracle server ....
Any idea ?
thanks in advance
jean-guy<HR></BLOCKQUOTE>
null

About data storing in SQL

Hi All
I've got one doubt recently Suppose
When we say <commit> to the insert /update/delete statements, the data will get stored in the data base permanently.
Before committing the statement where the data will be stored .

hi,
Theres are some background processes,SGA structures and datbasefiles which are used to process SQL statements.
All the insert statement data is store in Database buffer chache which on commit is written to to data files by database write.
All update,Delete statment results are stored in redolog buffer and the Log writer writen on commit to the redo log files.
The system monitor checks for consitency issues and if necessary initiates recovey of the database when the database is opened.
the check point process is responsible for updating database status information in the controlfiles and the datafiles when ever changes in the buffer chache are permanently recorded in the database.
For further detailed information see this link:
http://download-west.oracle.com/docs/cd/B12037_01/server.101/b10742/instance.htm#sthref203
Trinath Somanchi,
Hyderabad.

SAP GoLive : File System Response Times and Online Redologs design

Hello,
A SAP Going Live Verification session has just been performed on our SAP Production environnement.
SAP ECC6
Oracle 10.2.0.2
Solaris 10
As usual, we received database configuration instructions, but I'm a little bit skeptical about two of them :
1/
We have been told that our file system read response times "do not meet the standard requirements"
The following datafile has ben considered having a too high average read time per block.
File name -Blocks read - Avg. read time (ms) -Total read time per datafile (ms)
/oracle/PMA/sapdata5/sr3700_10/sr3700.data10 67534 23 1553282
I'm surprised that an average read time of 23ms is considered a high value. What are exactly those "standard requirements" ?
2/
We have been asked to increase the size of the online redo logs which are already quite large (54Mb).
Actually we have BW loading that generates "Chekpoint not comlete" message every night.
I've read in sap note 79341 that :
"The disadvantage of big redo log files is the lower checkpoint frequency and the longer time Oracle needs for an instance recovery."
Frankly, I have problems undertanding this sentence.
Frequent checkpoints means more redo log file switches, means more archive redo log files generated. right ?
But how is it that frequent chekpoints should decrease the time necessary for recovery ?
Thank you.
Any useful help would be appreciated.

Hello
>> I'm surprised that an average read time of 23ms is considered a high value. What are exactly those "standard requirements" ?
The recommended ("standard") values are published at the end of sapnote #322896.
23 ms seems really a little bit high to me - for example we have round about 4 to 6 ms on our productive system (with SAN storage).
>> Frequent checkpoints means more redo log file switches, means more archive redo log files generated. right?
Correct.
>> But how is it that frequent chekpoints should decrease the time necessary for recovery ?
A checkpoint is occured on every logswitch (of the online redologfiles). On a checkpoint event the following 3 things are happening in an oracle database:
Every dirty block in the buffer cache is written down to the datafiles
The latest SCN is written (updated) into the datafile header
The latest SCN is also written to the controlfiles
If your redologfiles are larger ... checkpoints are not happening so often and in this case the dirty buffers are not written down to the datafiles (in the case of no free space in the buffer cache is needed). So if your instance crashes you need to apply more redologs to the datafiles to be in a consistent state (roll forward). If you have smaller redologfiles more log switches are occured and so the SCNs in the data file headers (and the corresponding data) are closer to the newest SCN -> ergo the recovery is faster.
But this concept does not really fit the reality because of oracle implements some algorithm to reduce the workload for the DBWR in the case of a checkpoint.
There are also several parameters (depends on the oracle version) which control that a required recovery time is kept. (for example FAST_START_MTTR_TARGET)
Regards
Stefan

Sizing the redolog

My redolog details are below,
SQL> COLUMN value FORMAT 999,999,999,990
SQL> COLUMN name FORMAT a31
SQL> SELECT a.name,b.value
2 FROM v$statname a,v$sysstat b
3 WHERE a.statistic# = b.statistic#
4 AND a.name like '%redo%';
NAME VALUE
redo synch writes 281,602
redo synch time 27,625
redo blocks read for recovery 0
redo entries 24,978,833
redo size 9,970,044,216
redo buffer allocation retries 27,380
redo wastage 36,843,120
redo writer latching time 21
redo writes 141,083
redo blocks written 20,073,539
redo write time 346,493
NAME VALUE
redo log space requests 347,823
redo log space wait time 63,510
redo log switch interrupts 0
redo ordering marks 617,362
redo subscn max counts 740,460
16 rows selected.
SQL>
SQL>
I see that the redolog space request is high. I resized the redolog by 125m but still shows that the space request parameter has not changed. I dropped the redolog files and resized it to 125m. What i should do to reduce the space request close to 0?

Why increase log_buffer?There is a direct correlation between "redo log space requests" and a too-small log_buffer.
The Oracle Documentation notes:
"The redo log space requests Oracle metric indicates the active log file is full and Oracle is waiting for disk space to be allocated for the redo log entries. Space is created by performing a log switch.
Small Log files in relation to the size of the SGA or the commit rate of the work load can cause problems. When the log switch occurs, Oracle must ensure that all committed dirty buffers are written to disk before switching to a new log file. If you have a large SGA full of dirty buffers and small redo log files, a log switch must wait for DBWR to write dirty buffers to disk before continuing.
http://www.dba-oracle.com/t_log_buffer_optimal_size.htm
MetaLink note 216205.1 Database Initialization Parameters for Oracle Applications 11i, recommends a log_buffer size of 10 megabytes for Oracle Applications, a typical online database:
A value of 10MB for the log buffer is a reasonable value for Oracle Applications and it represents a balance between concurrent programs and online users.
The value of log_buffer must be a multiple of redo block size, normally 512 bytes.
Hope this helps. . .
Donald K. Burleson
Oracle Press author
Author of "Oracle Tuning: The Definitive Reference":
http://www.dba-oracle.com/bp/s_oracle_tuning_book.htm

REDO LOG BUFFER

Whenever a DML like Insert statement is issued it gets written to the Database buffer cache first by the server process(dedicated server).
Which process writes this DML activity to the redo log buffer ?
I guess DML is first written to the redolog files and only after that the same DML is committed to the data files.Is this correct ?
Can get any references to read on how any activity/DML is processed with a Oracle architecture perspective.
Thanks

Yes. Only the server process for that session knows what changes were made to the buffer cache. So it is the only one that can write the change vectors to the redo log buffer.
Hemant K Chitale

Log Buffer to disk?

Hi,
When the information in the Log Buffer is saved to disk?
Thanks,
Felipe

Hi Felipe,
Let’s conduct a brief summary about the redo process. When Oracle blocks are changed, including undo blocks, oracle records the changes in a form of vector changes which are referred to as redo entries or redo records. The changes are written by the server process to the redo log buffer in the SGA. The redo log buffer is then flushed into the online redo logs in near real time fashion by the log writer LGWR.
The redo logs are written by the LGWR when:
•     When a user issue a commit.
•     When the Log Buffer is 1/3 full.
•     When the amount of redo entries is 1MB.
•     Every three seconds
•     When a database checkpoint takes place. The redo entries are written before the checkpoint to ensure recoverability.
Remember that redo logs heavily influence database performance because a commit cannot complete until the transaction information has been written to the logs. You must place your redo log files on your fastest disks served by your fastest controllers. If possible, do not place any other database files on the same disks as your redo log files. Because only one group is written to at a given time, there is no harm in having members from several groups on the same disk.
To avoid losing information that could be required to recover the database at some point, Oracle has an archive (ARCn) background process that archives redo log files when they become filled. However, it’s important to note not all Oracle Databases will have the archive process enabled. An instance with archiving enabled, is said to operate in ARCHIVELOG mode and an instance with archiving disabled is said to operate in NO ARCHIVELOG mode.
You can determine with mode or if archiving is been used in your instance either by checking the value for the LOG_ARCHIVE_START parameter in your instance startup parameter file (pfile or spfile – This parameter is deprecated on version 10g), by issuing an SQL query to the v$database (“ARCHIVELOG” indicates archiving is enabled, and “NOARCHIVELOG” indicates that archiving is not enabled) or by issuing the SQL ARCHIVE LOG LIST command.
SQL> Select log_mode from v$database;
LOG_MODE
ARCHIVELOG
SQL> archive log list
Database log mode                    Archive Mode
Automatic archival                    Enabled
Archive destination                   USE_DB_RECOVERY_FILE_DEST
Oldest online log sequence       8
Next log sequence to archive   10
Current log sequence               10The purpose of redo generation is to ensure recoverability. This is the reason why, Oracle does not give the DBA a lot of control over redo generation. If the instance crashes, then all the changes within SGA will be lost. Oracle will then use the redo entries in the online redo files to bring the database to a consistent state. The cost of maintaining the redolog records is an expensive operation involving latch management operations (CPU) and frequent write access to the redolog files (I/O).
Regards,
Francisco Munoz Alvarez
www.oraclenz.com

Redo log and buffer size

Hi,
i'm trying to size redolog and the buffer size in the best way.
I already adjusted the size of the redo to let them switch 1/2 times per hour.
The next step is to modify the redo buffer to avoid waits.
Actually this query gives me 896 as result.
SELECT NAME, VALUE
FROM V$SYSSTAT
WHERE NAME = 'redo buffer allocation retries';
I suppose this should be near to 0.
Log_buffer is setted to 1m.
And i read "sizing the log buffer larger than 1M does not provide any performance benefit" so what can i do to reduce that wait time?
Any ideas or suggestions?
Thanks
Acr

ACR80,
Every time you create a redo entry, you have to allocate space to copy it into the redo buffer. You've had 588 allocation retries in 46M entries. That's "close to zero"..
redo entries 46,901,591
redo buffer allocation retries 588The 1MB limit was based around the idea that a large buffer could allow a lot of log to accumulate between writes with the result that a process could execute a small transaction and commit - and have to wait a "long" time for the log writer to do a big write.
If in doubt, check the two wait events:
"log file sync"
"log buffer space".
As a guideline, you may reduce waits for "log buffer space" by increasing the size of the log buffer - but this introduces the risk of longer waits for "log file sync". Conversely, reducing the log buffer size may reduce the impact of "log file sync" waits but runs the risk of increasing "log buffer space" waits.
Regards
Jonathan Lewis
http://jonathanlewis.wordpress.com
http://www.jlcomp.demon.co.uk

Redolog doubt in RAC

Hi,
I have small doubt regarding redolog corruption/deletion in RAC Instance:
OS: RHEL 4.5
Oracle: 10.2.0.1
DB_NAME=RAC ( 2 Instances, RAC1 and RAC2 are pointed to this DB).
One user (User1) is connected to DB using RAC1 Instance having redolog groups 2 (having each one member: redo1a.log and redo2a.log). I did logswitch freshly and Presently current redolog is redo2a.log. User1 created emp table and inserted 20 records and commited. note that, still current redolog is redo2a.log and there is no log switch. There are no users connected to DB other than user1.
Now, I shutdown abort the RAC1 instance and User1 session is switched to RAC2 Instance. When he queries the select * from emp table, he is getting 20 records? how come he gets? the inserted records are still in redolog2a.log of RAC1 instance and commit statement is also still there in redolog and thre is no log switch, instance RAC1 is shutdown abruptly. How come he gets 20 records, when he connected to RAC2 instance?
Please explain me, how it is possible? please execuse me, if there is anything wrong with my question.

Whether it is RAC or non RAC system DBWn process writes the dirty buffers to disk.
The DBWn process writes dirty buffer to disk under the following conditions:
1. When a checkpoint is issued.
2. When a server process cannot find a clean reusable buffer after scanning a threshold number of buffers.
3. Every 3 seconds
if still not written to disk when you start database next time media recovery will be completed.
Where in RAC environment it works on concept of GLOBAL CACHE. Whichever changes happening to 1st node same will be update to 2nd node as well. when node 1 goes down abruptly 2nd node identifies it and does the media recovery. so even when user again connects to 2nd node still data will be available.
Anil Malkai

Redologs on ASM but not archivelogs

Hello,
for the same database, is correct to have redolog on ASM when Archivelogs still on filesystem ?
Thank you.

Hello,
There is no correct answer to this problem. It depends on your strategy. The main advantages of ASM are simplicity and unbuffered I/O so make use of them.
Redo log need both write and read operations aside from archiving (bulk read). On the other side Archived Redo rarely need read operations after its initial bulk write, unless you do frequent recovery. So, you don't want to buffer archived redo logs.
Both can take advantage of ASM feature, so you can put both on ASM.
Later on, if are not satisfy with its performance, you can add more disks to ASM. It is easier to manage than putting archived redo logs on file system and the rest on ASM.
In my opinion it is more important to define strategy of storage tiers within the ASM itself. You can then map each tier to ASM disk group. For example for fast read/normal write you can create tier1 with RAID 1 or RAID 10 and map this tier to ASM disk group. You can then put OLTP/DSS data files into this tier (because they requires fast read). You can create other tiers according to your need based on this example.
clydie.com

Using RealPlayer (dialup) - How can I give N/W preference to it's buffer?

4-25-06
Is there a "safe" way to configure the machine so that "1st preference"
is given to keeping the RealPlayer buffer full (listening to radio stream)?
I use 56Kb dialup and don't mind slowing down e.g. Safari etc but don't
want the radio stream to be interrupted.
Have UDP ports 7070-7071 opened in Sys Pref's and tell RealPlayer to
use them. I also buffer 500 seconds in RealPlayer. Still, it's buffer gets
empty (stream is at 20+ Kbps and modem is at about 45+ Kbps).
Maybe there is a better forum to post this at discussions?
eMac, 10.3.9, RealPlayer 10 for Mac OS X, 10.0.0 (v331)
eMac G4 (build 7W98) 125 GHz, 256 MB, 37 GB Mac OS X (10.3.9) dialup 56K modem

just cleaning up my junk

Issue with running QuickTime Windows. Buffer Overrun Error - C++ Library .

Initial problem was Buffer Overrun Error (C++ Library) when clicking on QuickTime after installation. IE. QT would not even open. http://support.microsoft.com/kb/831875#appliesto
I took these steps:
1. Tried to uninstall QuickTime by itself (it failed).
2.
3. Manually deleted apple, itunes and quicktime from the entire system (where ever it let me).
4.
5. Manually took out from the registration the apple stuff.
6.
7. Left the items in the recycle bin (in case there were any real issues and I needed something restored).
8.
9. Performed a registration cure (RegCure).
10.
11. Took off my entire Anti Virus.
12.
13. Dropped down separately the QuickTime and separately the iTunes on desktop.
14.
15. Tried to install QuickTime from one of the two saved files on my desktop, but encountered a serious fault
16.
17. It needed the QuickTime Installer to remove QuickTime itself, else it crapped out and nothing happened. This complained for a QuickTime.msi file which was a problem.
18.
19. Went to the recycle bin and restored only components which were marked QuickTime Installer.
Removed the QuickTime instead of Repair.
Went to the website and Installed QuickTime 7 directly.
It opened on the desktop after installation.
Installed iTunes separately from the desktop and it opened directly.
Rebooted my pc.
Enabled all my Security (McAFee).
Opened one by one the QuickTime and Then the iTunes.
Created a computer restore point with a narrative for the future.
This was a very difficult task and required a lot of steps. I am glad you helped me with the removing part. Its great to have everything working again on my pc.
I hope this was helpful - it took me ageges to fix.

I'm experiencing exactly the same bug. Matter of fact, it's the first time in years that I've run across this kind of 'problem' when using non-beta software from a major player. Too bad. This really reflects poorly on Apple's credibility.

Redolog Buffer

Similar Messages

Maybe you are looking for