Instance crash on RAC will impact goldengate extract process?

Dear All,
DB Version: 11.2.0.3
GG version: 11.2.0.1
Could you please let me know, how gg works when a instance crashes in a RAC cluster. Will extract proess get abend or extract will continue to capture
Thank you very much in advance
Regards
Mvk

If GG is configured properly for the RAC, the GG manager should start extract on the living node and there should not be any data loss.
Please refer to Oracle documentation on how to configure GG in RAC.

Similar Messages

GoldenGate Extract Process will not read from redo log with manual help

Here is my issue.
I have GoldenGate replication successfully setup one-way from 1 Source to Many Targets. There is 1 source extract on the DB and many pumps that push the trail file data to the Targets. Replication does work but after manual help with starting up the Source Extract process.
If I execute the command:
GGSCI> alter extract <source extract name> begin now
GGSCI> view report <source extract name>
The extract starts and reads the source trail file but will not process data, I continually see in the ggserr.log file "OGG-01515: Positioning to begin time MMM DD, YYYY, HH:MM:SS" The date and time are irrevelant for this problem.
When I see this, I SQ*Plus into the database and look in the v$log table for the current log and sequence #.
I return to GGSCI and issue the following command:
GGSCI> alter extract <source extract name> thread 1 extseqno <sequence # from v$log query>
GGSCI> start <source extract name>
It then works as expected. Why is this so? I thought the alter extract <source extract name> begin now would do the same output.
We do use ASM but like I said when I issue the:
GGSCI> alter extract <source extract name> thread 1 extseqno <sequence # from v$log query>
It works like it should.
Very weird.
- Jason

Yes, supplemental logging is enabled on both the source and the targets, but why would supplemental logging on the targets have any affect on why the Source Extract on the source can't read from the source redo log?
This is not a RAC database, rather single-instance with one thread. Also, we are using DBLOGREADER functionality as it is an 11.2.0.3 database.
My issue is simply, when I start the source extract from being down, meaning it isn't running, I issue this command:
alter <source extract> begin now
start <source extract>
view report <source extract>
OGG-01515 Positioning to begin time <today's date and time> ie Mar 4, 2013, 3:26:39 PM. (this is repeated over and over and over)
If I perform a
info <source extract> detail---> I see the following:
Log Read Checkpoint Oracle Redo Logs 2013-03-04 15:26:39 Thread 1, Seqno 0, RBA 0 (why is it showing 0, becuase it can't read the redo, WHY NOT?)
Extract Source BEGIN END
Not Available <today's date> <today's date> (repeat....)
However, if I retreive the Redo Log number and I issue:
alter spe thread 1 extseqno (redo log sequence #)
start spe.
Then it works okay. I have to manually tell it what redo log to begin reading from. Why?
- Jason
Edited by: 924317 on Mar 4, 2013 9:03 AM

OC4J_SOA INSTANCE CRASH INADVERTENTLY WHILE A ESB SYSTEM IS PROCESSING

We are using OAS 10.1.3.3.0....
We have a ESB system which reads changes in a table (DB Adapter) and enqueue that changes in an AQ queue with an ADT (AQ Adapter)...
Sometimes that ESB process thousands of records very well but sometimes when is processing the oc4j_soa instances
crash and its log show...
07/11/08 17:03:57 at java.lang.Thread.run(Thread.java:595)
# An unexpected error has been detected by HotSpot Virtual Machine:
# SIGSEGV (0xb) at pc=0xb7a0010c, pid=28819, tid=831085488
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# V [libjvm.so+0x3d910c]
# An error report file with more information is saved as hs_err_pid28819.log
# If you would like to submit a bug report, please visit:
# http://java.sun.com/webapps/bugreport/crash.jsp
When I see the AppServer Control the instance is automatically restarted....
I need your help......!
Message was edited by:
user604794

Thanks for your answer....
About it.. let me talk to you the following... I have experienced this crash in Windows XP in a development enviroment, and I think that in an enviroment more robust like Linux this problem woulbe be solved... but no... the same error with less frecuency but the same error...
About Metalink... they told me....
"...You should be able to log a bug with SUN. The JVM is a SUN product. From the error is looks like an issue with one
of the JVM libraries. Once you submit the hs_err_pid28819.log file, SUN should b
e able to search and find a match in there library regarding the issue...."
But In SUN when I try to register the crash I read the following...
" ...If the crash occurred while running a product developed by a company other than Sun, please report this issue to the vendor of that product. ..."
So I don´t know what to do... Anybody HELP....

Only one instance in a RAC will start at a time

All:
I have a 4 node Oracle RAC running on RHEL AS 4 64 bit. 10.2.0.1 64 Bit
This DB was running just fine last week.
Now all of a sudden only one instance will start.
In the alert log I get an ORA-1102
crsctl shows everything good
srvctl show all node apps running
srvctl shows all ASM instances running
When the second instance starts the alert log shows the reconfiguration taking place.
Then the second instance will die after getting an ORA-1102.
I have tried deleting and re-adding all the instances with DBCA.
ocrcheck comes back fine on all nodes.
I have another RAC on the same set of HP Blades that is still hunky dorey so I don't know what is going on.
Please help.

Hi,
check the value of cluster_database parameter
it should be TRUE on all instances, also post ur alert log and trace file.
Thanks
Kuljeet

Rac Instance Crashes

Dear all,
My version is 11.2.0.2.5 one of my rac instance crashes with message ORA-00240: control file enqueue held for more than 120 seconds. Received an instance abort message from instance 1.
here are the contents of alert log file
IPC Send timeout detected. Receiver ospid 27423 [[email protected] (LMON)]
2013-03-22 22:30:05.644000 -07:00
Errors in file /u01/app/oracle/diag/rdbms/lfgoimdb/LFGoimdb2/trace/LFGoimdb2_lmon_27423.trc:
2013-03-22 22:31:08.734000 -07:00
Errors in file /u01/app/oracle/diag/rdbms/lfgoimdb/LFGoimdb2/trace/LFGoimdb2_arc2_27691.trc (incident=15905):
ORA-00240: control file enqueue held for more than 120 seconds
Incident details in: /u01/app/oracle/diag/rdbms/lfgoimdb/LFGoimdb2/incident/incdir_15905/LFGoimdb2_arc2_27691_i15905.trc
2013-03-22 22:31:13.409000 -07:00
Received an instance abort message from instance 1
Please check instance 1 alert and LMON trace files for detail.
LMS0 (ospid: 27427): terminating the instance due to error 481
System state dump requested by (instance=2, osid=27427 (LMS0)), summary=[abnormal instance termination].
System State dumped to trace file /u01/app/oracle/diag/rdbms/lfgoimdb/LFGoimdb2/trace/LFGoimdb2_diag_27413.trc
2013-03-22 22:31:18.376000 -07:00
Dumping diagnostic data in directory=[cdmp_20130322223113], requested by (instance=2, osid=27427 (LMS0)), summary=[abnormal instance termination].
ORA-1092 : opitsk aborting process
Instance terminated by LMS0, pid = 27427

Thanks for reply,
My redo logs size is default 50mb.There is currently no load on the system since we are not using this environment for time being.The log switches are averaged to be 8 per day.I think Increasing the size of redo will further cause the problems since the archiver may again hold lock for more time.
Since there is no dedicated connection between the nodes and storage ,So increasing the hardware and network configuration is only solution to this? Or I am still missing something...
As far as configuration is considered i cannot add more resources to this environment.How can I solve this issue?

Difference between Instance Crash and Database Crash in Standalone and RAC?

I am new to Oracle Environment. I would appreiciate if i get clarified on the "_+Difference between Instance Crash and Database Crash in Standalone and
RAC_".+
Oracle Version: 11.1.0.7.0
O.S Version: HP-UX B.11.31 U

Sivaprasad S wrote:
I am new to Oracle Environment. I would appreiciate if i get clarified on the "_+Difference between Instance Crash and Database Crash in Standalone and
RAC_".+That depends on your definitions of database crash versus instance crash.
Literally, database crash means that the database itself is damaged in some way. A disk is no more. Data files are corrupted. Etc. The database itself thus causes the crash - and a database crash will also cause all database instances (RAC or single) for that database to crash too.
An instance crash will therefore mean that the crash occurs in the instance due to an Oracle bug or similar. In other words a s/w error and not a database error. The database is still "fine", but the instance not. Instance crashes. However, any other instance on that same (RAC) database will still be fine as the database itself is unaffected by that s/w error in the crashed instance.
If you mean something different, please enlighten us. Not all crashes are equal. {noformat};-){noformat}

How to restore a single instance database to RAC (10g) using RMAN ?

Hi all,
I have a single instance database configured using file system in a unique server, so now I need to restore this database using RMAN to a new RAC environment with two nodes and ASM also.
Does anyone can help with this documentation or a link about any information about this restore using rman ?
I tryied found on metalink but threres isn´t this specific case. Only how to convert single instance to 10g RAC, but without how to use with rman information.
System: Windows 2003
Oracle 10g R2
Thanks.
Wander(Brazil)

Wander(Brazil) wrote:
Hi all,
I have a single instance database configured using file system in a unique server, so now I need to restore this database using RMAN to a new RAC environment with two nodes and ASM also.
In essense you are really doing nothing more than restoring a backup on a new host. If you want a blow by blow detail of how to accomplish it take at http://www.oracledistilled.com/oracle-database/restore-database-to-another-host-using-rman
In order to write the files to the ASM Disk groups you will use the SET NEWNAME FOR DATAFILE in your restore script. You will also need to create the ORACLE_BASE/admin/<DBSID> structure on all of node. Once you get the database completely restore you will need to edit the SPFILE to include the proper value for the CLUSTER* parameters and each the node specific parameters. Sorry that it is kind of higher level than what you are probably looking for but the link above and what I put here should get you were want to go.

Instance naming in RAC environment

DB version : 11.1.0.7
OS : AIX 6.1
It has been few months since i've installed a RAC DB.
After the GRID/CRS installation, while DB creation, when you specify the DB name in DBCA , say orcl, DBCA will automatically create instances orcl1 and orcl2,...
But in my new company, i've come across RAC environments where the Instance names of the DB is not like
DB Name: ORCL
instance 1: orcl1
instance 2: orcl2Rather it is like
-- From Node1
SQL> select name from v$database;
NAME
MKESCU
SQL> select instance_name from v$instance;
INSTANCE_NAME
mkes1uHow did they manage to do this?

Why would anyone want instances for a RAC DB :MKESCU to be named as
mkes1u
mkes2uIs it for any security purpose?
Isn't
mkescu1
mkescu2straightforward and easy to understand?

Which file decide instance name in RAC 10g ?

Which file decide instance name in RAC 10g ?
How does RAC know which instance should start from which machine?
Thanks.

The spfile is NOT the correct answer, actually.
1. The instance name for any Oracle instance, whether in a RAC or not, is derived from the value for the ORACLE_SID parameter which is in force at the time the startup command is issued. If you were to:
export ORACLE_SID=SALES
sqlplus / as sysdba
startup...you would just have created an instance called SALES.
2. The spfile (or init.ora if you're not using an spfile) contains a parameter called INSTANCE_NAME. That defaults to whatever ORACLE_SID is set to as an environment variable when the instance is started, hence point 1 above. But you can force an instance to acquire a non-SID name by deliberately setting it to something different. That's unusual to do, though, except in a RAC context where you want to subvert the default load balancing mechanism.
2b. The spfile (or init.ora) also contains a parameter called DB_NAME -but that is the name of the database, not the name of the instance, and is therefore irrelevant for the purposes of working out how instances get their names,
3. Neither of the above answers has got anything to do with RAC knowing what instance runs on which machine. That is the job of the Oracle Cluster Registry file (OCR), for which you have to provide about 100MB of space on a shared hard drive during the RAC installation itself.
You populate that file by running DBCA to create your RAC instances (and DBCA then automatically updates the OCR to record the fact that it created instance X1 on Node A and X2 on Node B, and therefore that's what will happen in the future unless you change things).
Or you can update the OCR yourself by using the SRVCTL utility. You'd do something like: srvctl add instance -d DB1 -i X3 -n C (add an instance called X3 to run on Node C as part of database DB1).
A little more details about the OCR is available in Section 13 of this article, for example: http://www.dizwell.com/prod/node/26?page=0%2C5

Can we assign one Conc,Manager to any one instance of a RAC DB

Hi All.
Can we assign one concurrent request which is submitted by a particular Conc,Manager to any one instance of a RAC DB
Thanks

Does anyone have any documentation explaining how to do this? I've set up a new RAC service for batch processing -- I'd like some of my concurrent requests (all running through one Concurrent Manager) to use this RAC service. How do I tell a Concurrent Manager to use a specific RAC service? Does the Concurrent Manager have to point to a primary Application server (we also have PCP enabled), and then update the tnsnames.ora with the RAC service? I've tried that, but I still don't see how the request will be channeled through to the "batch01_svc" service. Thanks!

Oracle instance crashed with error 472

My Oracle instance crashed after running for 2 days
with 5 reads and 2 updates in every 4 seconds. The
alert log are
Fri Oct 24 02:59:13 2003
found dead multi-threaded server 's000',pid=(10,1)
Fri Oct 24 03:00:31 2003
LGWR: terminating instance due to error 472
Instance terminated by LGWR, pid = 109
The client access the data through stored procedures.
The database is running on Windows NT 4.0 service pack 6a.
The Oracle is v8.1.5.0.0
Any help will be greatly appreciated.

ORA-00472 PMON process terminated with error
Cause: The Process Monitor process terminated abnormally.
Action: Check the accompanying messages, and the background process trace file. Correct the problem mentioned in the messages. Then shut down and restart the instance. If the trace file mentions any other background process messages, check the trace file for the mentioned process until the root message is found.
Follow the recomendations above
Joel P�rez

Goldengate extracting from Archive log

Dear All,
We are conducting a replication of Oracle Database. Due to high load on Production server, the client is providing us a near DR database server for extraction. This database server is using Oracle Data Guard to be in Sync with Production server. Hence we will be using Archive logs shipped from Production to DR to Extract changed data. I have searched on the Net and read manuals, but I am unable to understand the how to configure OGG for this scenario. I understand that we can used Tranlogoptions parameter to extract data from Archive logs, but since the near DR database is in Mount state, how can we Log in the database using userid /password in extract process. Also we need to have Select any dictionary privileges on the database, but database is in mount state, so how to work on it??
Awaiting eagerly for the reply.l

For a physical standby, you can configure GG to read the archived standby logs in archive log only (ALO) mode.
ARCHIVEDLOGONLY causes Extract to read from the archived logs exclusively, without querying or validating the logs from system views such as v$log and v$archived_log. This parameter puts Extract into Archived Log Only mode (ALO). For more information, see the Oracle GoldenGate Oracle Installation and Setup Guide.
You can configure the Extract process to read exclusively from the archived logs. This is
known as Archived Log Only (ALO) mode. In this mode, Extract only reads from archived
logs that are stored in a specified location. ALO mode allows Oracle GoldenGate to use
production logs that are shipped over to a secondary database (such as a standby) as the
data source for Oracle GoldenGate. The online logs will not be used. Oracle GoldenGate
will connect to the secondary database to get metadata and other required data as needed.
As an alternative, ALO mode is supported on the production system.
Extract automatically operates in ALO mode if it detects that the database is a physical standby.
Lots of other info in the Oracle install guide.

OC4J instance crashing after Ping Failed: default_group~home~default_group~

Hi All,
I have noticed from opmn logs that [pm-process] Process Crashes and System has been restarted automatically by itself.
I have read the tread with title "OC4J instance crashing - can't find any diagnostics " and I have configured the opmn.xml file in order to enable opmn.dbg logs.
In the meantime I have suspected that the BPEL crashes are the same date with default_group~home~default_group logs rotation.
[oracle@csporacle1 logs] cat /etc/redhat-release
Red Hat Enterprise Linux ES release 4 (Nahant Update 5)
[oracle@csporacle1 logs] cat opmn.log | grep "Restarting"
10/05/06 17:10:18 [pm-process] Process Crashed: default_group~home~default_group~1 (481247799:27664) - Restarting
10/07/08 03:38:12 [pm-process] Process Crashed: default_group~home~default_group~1 (481247801:12964) - Restarting
10/08/05 15:01:08 [pm-process] Process Crashed: default_group~home~default_group~1 (481247802:22304) - Restarting
10/09/10 04:19:56 [pm-process] Process Crashed: default_group~home~default_group~1 (481247803:5081) - Restarting
10/10/15 00:04:21 [pm-process] Process Crashed: default_group~home~default_group~1 (481247804:7275) - Restarting
10/11/20 14:56:29 [pm-process] Process Crashed: default_group~home~default_group~1 (481247806:28469) - Restarting
10/12/25 11:15:39 [pm-process] Process Crashed: default_group~home~default_group~1 (481247807:17717) - Restarting
11/02/02 22:09:16 [pm-process] Process Crashed: default_group~home~default_group~1 (481247808:540) - Restarting
11/02/26 12:14:55 [pm-process] Process Crashed: default_group~home~default_group~1 (481247809:26490) - Restarting
11/03/12 04:28:19 [pm-process] Process Crashed: default_group~home~default_group~1 (481247812:7348) - Restarting
11/04/25 05:43:59 [pm-process] Process Crashed: default_group~home~default_group~1 (481247817:9090) - Restarting
11/05/21 08:27:55 [pm-process] Process Crashed: default_group~home~default_group~1 (481247818:26304) - Restarting
11/06/18 08:04:03 [pm-process] Process Crashed: default_group~home~default_group~1 (481247826:28958) - Restarting
11/07/12 19:04:46 [pm-process] Process Crashed: default_group~home~default_group~1 (481247828:10867) - Restarting
11/08/21 14:33:08 [pm-process] Process Crashed: default_group~home~default_group~1 (481247832:13556) - Restarting
11/09/18 20:40:16 [pm-process] Process Crashed: default_group~home~default_group~1 (481247835:22024) - Restarting
11/10/17 13:56:47 [pm-process] Process Crashed: default_group~home~default_group~1 (481247838:20504) - Restarting
[oracle@csporacle1 logs] ls -lrt
total 6020344
-rw-r--r-- 1 oracle dba          0 Feb 17 2010 opmn.out
-rw-r--r-- 1 oracle dba   84895911 Feb 2 2011 default_group~home~default_group~1.log.11-02-02_22:09:16.gz
-rw-r--r-- 1 oracle dba   83151884 Feb 26 2011 default_group~home~default_group~1.log.11-02-26_12:14:55.gz
-rw-r--r-- 1 oracle dba   79803990 Mar 12 2011 default_group~home~default_group~1.log.11-03-12_04:28:19.gz
-rw-r--r-- 1 oracle dba   66470738 Mar 29 2011 default_group~home~default_group~1.log.11-03-29_00:17:05.gz
-rw-r--r-- 1 oracle dba   84542865 Apr 25 2011 default_group~home~default_group~1.log.11-04-25_05:43:59.gz
-rw-r--r-- 1 oracle dba   83672948 May 21 08:27 default_group~home~default_group~1.log.11-05-21_08:27:55.gz
-rw-r--r-- 1 oracle dba   84825282 Jun 18 08:04 default_group~home~default_group~1.log.11-06-18_08:04:03.gz
-rw-r--r-- 1 oracle dba   86740423 Jul 12 19:04 default_group~home~default_group~1.log.11-07-12_19:04:46.gz
-rw-r--r-- 1 oracle dba   66921060 Aug 3 23:06 default_group~home~default_group~1.log.11-08-04_02:28:06.gz
-rw-r--r-- 1 oracle dba   71221025 Aug 21 14:33 default_group~home~default_group~1.log.11-08-21_14:33:08.gz
-rw-r--r-- 1 oracle dba 2147483647 Sep 18 20:39 default_group~home~default_group~1.log.11-09-18_20:40:16
-rw-r--r-- 1 oracle dba 2147483647 Oct 17 13:56 default_group~home~default_group~1.log.11-10-17_13:56:49
drwxr-x--- 2 oracle dba       4096 Oct 24 00:11 states
-rw-r--r-- 1 oracle dba      64203 Oct 24 20:09 opmn.log
-rw-r--r-- 1 oracle dba 976001440 Oct 28 02:35 default_group~home~default_group~1.log
-rw-r--r-- 1 oracle dba   95431960 Oct 28 02:36 opmn.dbg
[oracle@csporacle1 logs] After this I have check the opmn.xml and found that the ORACLE_HOME has been set to other oracle on the same Unit.
oracle@csporacle1 opmn] cat conf/opmn.xml | head -20
<?xml version = '1.0' encoding = 'UTF-8'?>
<opmn xmlns="http://www.oracle.com/ias-instance">
   <log path="$ORACLE_HOME/opmn/logs/opmn.log" comp="internal;ons;pm" rotation-size="1500000"/>
   <debug path="$ORACLE_HOME/opmn/logs/opmn.dbg" comp="internal;pm;ons" rotation-size="1500000"/>
   <notification-server interface="ipv4">
      <port local="6100" remote="6200" request="6003"/>
      <ssl enabled="true" wallet-file="$ORACLE_HOME/opmn/conf/ssl.wlt/default"/>
   </notification-server>
   <process-manager>
      <process-modules>
         <module path="$ORACLE_HOME/opmn/lib/libopmnohs">
            <module-id id="OHS"/>
         </module>
         <module path="$ORACLE_HOME/opmn/lib/libopmnoc4j">
            <module-id id="OC4J"/>
         </module>
         <module path="$ORACLE_HOME/opmn/lib/libopmncustom">
            <module-id id="CUSTOM"/>
         </module>
         <module path="$ORACLE_HOME/opmn/lib/libopmniaspt">
[oracle@csporacle1 opmn] pwd
/space/u01/app/oracle/product/10.1.3.1/OracleAS_1/opmn
[oracle@csporacle1 opmn] cd $ORACLE_HOME
[oracle@csporacle1 10.1] pwd
/u01/app/oracle/product/10.1
[oracle@csporacle1 10.1] Are these error are related with my findings or are there any criteria must i check further.
Could you please help me to solve this problem.
Thanks & Best Regards,
Bahtiyar

gday mattysmith --
Unfortunately that's a very good question.
As OPMN is in the picture here, the first step may be to enable debug logging for OPMN in the chance that it may give you more information about why it thinks the process crashed.
Generally speaking, when OPMN is restarting a process, it can be because it recognized that a managed process has dissapeared (ie JVM core dumped, etc.) or it can also be because the JVM didn't respond to any of OPMNs requests to let it know it was still alive. In which case OPMN will kill then restart the managed process. This can be caused by situations where the JVM has run out of memory, etc. and becomes unresponsive, leading to it eventually being detected as such and restarted.
What you can do is to add the OPMN PM submodule to the OPMN debug logger.
<opmn xmlns="http://www.oracle.com/ias-instance">
<log path="$ORACLE_HOME\opmn\logs\opmn.log" comp="internal;ons;pm" rotation-s
ize="1500000"/>
<debug path="$ORACLE_HOME\opmn\logs\opmn.dbg" comp="internal;pm;ons" rotation-size="1500000"/>
</opmn>
That should then produce an opmn.dbg file in the $ORACLE_HOME/opmn/logs directory. It'll be very verbose, so you'll need to associate the restart times from the other log file to locate the relevant section of the file. Hopefully that may shed some light on what is happening, at least at the restart level.
Usually if there are runtime exceptions such as OOM errors, etc. then those are dumped to stderr/stdout. Check the contents of the other log files in the opmn/logs directory of the form default_group~home~default_group~1.log to see if there is anything obvious in there.
There has been reports of an issue with PermGen space usage -- that could be something you are running into under the covers. Try resizing the permgen space to be a little larger and see if the problem goes away -- that can be done in opmn.xml in the java > start-options section of the instance in question.
See how any of that goes!
-steve-

Goldengate Extracts reads slow during Table Data Archiving and Index Rebuilding Operations.

We have configured OGG on a near-DR server. The extracts are configured to work in ALO Mode.
During the day, extracts work as expected and are in sync. But during any dialy maintenance task, the extracts starts lagging, and read the same archives very slow.
This usually happens during Table Data Archiving (DELETE from prod tables, INSERT into history tables) and during Index Rebuilding on those tables.
Points to be noted:
1) The Tables on which Archiving is done and whose Indexes are rebuilt are not captured by GoldenGate Extract.
2) The extracts are configured to capture DML opeartions. Only INSERT and UPDATE operations are captured, DELETES are ignored by the extracts. Also DDL extraction is not configured.
3) There is no connection to PROD or DR Database
4) System functions normally all the time, but just during table data archiving and index rebuild it starts lagging.
Q 1. As mentioned above, even though the tables are not a part of capture, the extracts lags ? What are the possible reasons for the lag ?
Q 2. I understand that Index Rebuild is a DDL operation, then too it induces a lag into the system. how ?
Q 3. We have been trying to find a way to overcome the lag, which ideally shouldn't have arised. Is there any extract parameter or some work around for this situation ?

Hi Nick.W,
The amount of redo logs generated is huge. Approximately 200-250 GB in 45-60 minutes.
I agree that the extract has to parse the extra object-id's. During the day, there is a redo switch every 2-3 minutes. The source is a 3-Node RAC. So approximately, 80-90 archives generated in an hour.
The reason to mention this was, that while reading these archives also, the extract would be parsing extra Object ID's, as we are capturing data only for 3 tables. The effect of parsing extract object id's should have been seen during the day also. The reason being archive size is same, amount of data is same, the number of records to be scanned is same.
The extract slows down and read at half the speed. If normally it would take 45-50 secs to read an archive log of normal day functioning, then it would take approx 90-100 secs to read the archives of the mentioned activities.
Regarding the 3rd point,
a. The extract is a classic extract, the archived logs are on local file system. No ASM, NO SAN/NAS.
b. We have added "TRANLOGOPTIONS BUFSIZE" parameter in our extract. We'll update as soon as we see any kind of improvements.

Instance crash?

Imagine these circumstances:
1. I make an update to one/multiple rows of a table, but do not commit
2. My instance crashes.
How does Oracle manage this? Please correct me if i am wrong
Below is what i have understood
1. SMON does the instance Recovery (Rolling Forward only) for below cases
     a)Uncommited transaction will be recovered if only dirty buffer is written to the datafile even not committed [ alter system checkpoint ]
     b)Uncommited transaction will be recovered if only Redo is generated or log switches occur, or the logs are archived [alter system switch
       logfile/alter system archive log current]
2. Uncommited transaction will be not be recovered if all above of these doesn't happen.Then SMON will do [ Rolling Forward + Rolling Backward ]

913410 wrote:
Imagine these circumstances:
1. I make an update to one/multiple rows of a table, but do not commit
2. My instance crashes.
How does Oracle manage this? Please correct me if i am wrong
Below is what i have understood
1. SMON does the instance Recovery (Rolling Forward only) for below cases
     a)Uncommited transaction will be recovered if only dirty buffer is written to the datafile even not committed [ alter system checkpoint ]
     b)Uncommited transaction will be recovered if only Redo is generated or log switches occur, or the logs are archived [alter system switch
       logfile/alter system archive log current]
2. Uncommited transaction will be not be recovered if all above of these doesn't happen.Then SMON will do [ Rolling Forward + Rolling Backward ]
This is quite well explained here,
http://docs.oracle.com/cd/E11882_01/server.112/e25789/startup.htm#CNCPT1301
Aman....

Instance crash on RAC will impact goldengate extract process?

Similar Messages

Maybe you are looking for