AWR & ADDR Report

Hi ,
any one know how we cane read AWR report any basedline to comper ?? below some resuilts how can we now it's good or not .
Finding 2: Commits and Rollbacks
Impact is .02 active sessions, 10.04% of total activity.
Waits on event "log file sync" while performing COMMIT and ROLLBACK operations
were consuming significant database time.
Recommendation 1: Application Analysis
Estimated benefit is .02 active sessions, 10.04% of total activity.
Action
Investigate application logic for possible reduction in the number of
COMMIT operations by increasing the size of transactions.
Rationale
The application was performing 683 transactions per minute with an
average redo size of 4638 bytes per transaction.

Hello,
A Log file sync occurs when a process waits for the LGWR to confirm that the their redo has been flushed from the redo log buffer to disk.
The most common causes for this are:
1. Slow I/O subsystem on the location where the Redo Log files are stored.
2. Execessive commits. (i.e. an application inserting a lot of rows and commiting after each one)
3. Excessive switches.
Log file parallel write is the event waited for the actual write operation that the LGWR is perfoming from the log buffer to the disk. You should compare your average MS of Log File Parallel Write to yoiur Log File Sync event, and see how much of the log file Sync is actually waited on I/O. If it is a big part, then check your devices for slow I/O, and/or concurrency.
Check how many redo log switch per hour you are doing, It's possible that the increased DML activity from your load test is provoking an increse in redo switch activity due to your redolog file size. A redolog switch, ideally should not occur more often than once every 15 minutes.
Lastly, check if your application is commiting too agressively. This is likely your case. This can be derived in your AWR through your user calls/ commit ratio. At your Instance Activity part of the report check for the statistics:
user calls
user commits
user rollbacks
And then do : user calls/(user commits+user rollbacks)
This value should above 20-25, if not, your application is commiting too frequently.

Similar Messages

Cannot generate AWR/ADDM reports in Oracle 10g

Hi,
We are running 10.2.0.3.0 and troublshooted the performance problem for some queries. But when tried to run AWR/ADDM reports from OEM and got the following messages:
Insufficient Data in Interval. For displaying data on this page, two historical snapshots are needed. Make sure that two snapshots are present in the target database instance. In addition modify the interval so that it is contained within two available snapshots
I checked the scheduled jobs in this DB,
Connected to:
Oracle Database 10g Enterprise Edition Release 10.2.0.3.0 - 64bit Production
With the Partitioning, Real Application Clusters, Oracle Label Security and Data Mining options
SQL> select JOB_NAME from dba_scheduler_jobs;
JOB_NAME
PURGE_LOG
FGR$AUTOPURGE_JOB
GATHER_STATS_JOB
AUTO_SPACE_ADVISOR_JOB
Has anyone anyidea why AWR/ADDM reports cannot be run? Thanks
Liz

Hi,
If you want you can execute snapshot at your required time as suggested by people.
However Oracle 10g by default take snap at every one hour.
To check
select snap_id, to_char(BEGIN_INTERVAL_TIME,'DD-MON-YYYY HH24:MI') BEGIN_INTERVAL_TIME,
to_char(END_INTERVAL_TIME,'DD-MON-YYYY HH24:MI') END_INTERVAL_TIME
from dba_hist_snapshot
You can generate AWR Report by executing following SQL, Connect as SYS
SQL> @$ORACLE_HOME/rdbms/admin/awrrpt.sql
Input required details it asks, you can get AWR report in HTML or text format.
ADDM Report:
@$ORACLE_HOME/rdbms/admin/addmrpt.sql

Import AWR Html report into another database

Hello,
Im in 11gR2, is it possible to import the AWR Html report from database 1 into another database 2?
I would like to use ADDM from another database 2 to analyze the AWR Html report of database 1.
Thanks,

AWR report importing on another database
i dont think so

How can i save a AWR/statspack report in xml format ??

how can i save a AWR/statspack report in xml format ?? or is there a way...coz when creating the report it gives option for html or normal (sql or txt based file)... on 10.2.0.3

thanks orinet....i used wat you asked me to use...it generated a .dmp file....
sqlplus "/as sysdba"
SQL> @$ORACLE_HOME/rdbms/admin/awrextr.sql
i tired upload that file to orapert.com it gives me below error ...any reason why ??
Error : File not processed because it was too small to be an Oracle statspack or bstat/estat file.
do i need to load it up too ?? as i tried doing the below
sqlplus "/as sysdba"
SQL>@$ORACLE_HOME/rdbms/admin/awrload.sql
get error while doing that....
ERROR at line 1:
ORA-20105: unable to move AWR data to SYS
ORA-06512: at "SYS.DBMS_SWRF_INTERNAL", line 1760
ORA-20107: not allowed to move AWR data for local dbid
ORA-06512: at line 3
the file is about 19M in size
Edited by: user630084 on Mar 19, 2009 11:37 AM

Statspack, awr, addm reports

Hi every one,
what is the difference between statspack, awr, addm reports and how these report are going to analyze and how these reports are going to use in real time.
please let me know.
thank ypu!

851707 wrote:
Hi every one,
what is the difference between statspack, awr, addm reports and how these report are going to analyze and how these reports are going to use in real time.
The very major difference between Statspack and the other reports is that the former one is a free one and all the rest, require a licence before you can legally use them. For the usage of both the types of reports, please read the entire series for the same written by Jonathan Lewis on http://jonathanlewis.wordpress.com/statspack-examples/
HTH
Aman....

Automating AWR & ADDM Reports

Just wanted to check, if anybody has done... Automatic Generation of AWR & ADDM report. I want send the report to all the dba's in the team every six hours... Is there a way to call through cron Job....Want to pass parameter last hours snapshot id and snapshot_id -6...
Only if you have done...pl. answer otherwise ignore...
Thanks

select output from
table(dbms_workload_repository.&fn_name( :dbid,
inst_num,
bid, :eid,
rpt_options ));
Figure out what you need from there. Wrap spool
around it for your directory, mail it off when done.nice little script Steven.
Niall Litchfield
http://www.orawin.info/

Is there an AWR fragmentation report I can run manually

I am using 10g on Redhat Linux. I am not using Oracle grid or dbconsole.
Are ther any AWR reports that I can run manually from the Linux command line or SQLPLUS that will show me all tables with heavy fragmentation?
Thanks in advance.

variable id number;
begin
declare
name varchar2(100);
descr varchar2(500);
obj_id number;
begin
name:='Manual_Employees';
descr:='Segment Advisor Example';
dbms_advisor.create_task (
advisor_name => 'Segment Advisor',
task_id => :id,
task_name => name,
task_desc => descr);
dbms_advisor.create_object (
task_name => name,
object_type => 'TABLE',
attr1 => 'HR',
attr2 => 'EMPLOYEES',
attr3 => NULL,
attr4 => NULL,
attr5 => NULL,
object_id => obj_id);
dbms_advisor.set_task_parameter(
task_name => name,
parameter => 'recommend_all',
value => 'TRUE');
dbms_advisor.execute_task(name);
end;
end;
Viewing Segment Advisor Results
The Segment Advisor creates several types of results: recommendations, findings, actions, and objects. You can view results in the following ways:
•With Enterprise Manager
•By querying the DBA_ADVISOR_* views
•By calling the DBMS_SPACE.ASA_RECOMMENDATIONS procedure
http://download.oracle.com/docs/cd/B28359_01/server.111/b28310/schema003.htm
Kind Regards,
Rakesh

AWR sql report

Hi,
on 11g R2, I ran awrsqrpt.sql and received the following :
-> % Total DB Time is the Elapsed Time of the SQL statement divided
   into the Total Database Time multiplied by 100
Stat Name                                Statement   Per Execution % Snap
Elapsed Time (ms)                         2,483,336    2,483,335.9     1.2
CPU Time (ms)                               138,340      138,340.0     0.2
Executions                                        1            N/A     N/A
Buffer Gets                               2,176,636    2,176,636.0     0.1
Disk Reads                                  805,870      805,870.0     0.6
Parse Calls                                       1            1.0     0.0
Rows                                        137,462      137,462.0     N/A
User I/O Wait Time (ms)                   2,265,412            N/A     N/A
Cluster Wait Time (ms)                      123,124            N/A     N/A
Application Wait Time (ms)                        0            N/A     N/A
Concurrency Wait Time (ms)                        0            N/A     N/A
Invalidations                                     0            N/A     N/A
Version Count                                     3            N/A     N/A
Sharable Mem(KB)                                239            N/A     N/A
Execution Plan
| Id | Operation                       | Name               | Rows | Bytes | Cost (%CPU)| Time     |
|   0 | INSERT STATEMENT                |                    |       |       |   441K(100)|          |
|   1 | LOAD TABLE CONVENTIONAL        |                    |       |       |            |          |
|   2 |   NESTED LOOPS                  |                    |       |       |            |          |
|   3 |    NESTED LOOPS                 |                    |     1 |   172 |   441K (1)| 01:28:19 |
|   4 |     NESTED LOOPS                |                    |     1 |   135 |   441K (1)| 01:28:19 |
|   5 |      TABLE ACCESS BY INDEX ROWID| PS_KK_ACTIVITY_LOG |     1 |   105 |   441K (1)| 01:28:19 |
|   6 |       INDEX RANGE SCAN          | PSEKK_ACTIVITY_LOG | 1982K|       | 9418   (1)| 00:01:54 |
|   7 |      TABLE ACCESS BY INDEX ROWID| PS_KK_SOURCE_HDR   |     1 |    30 |     2   (0)| 00:00:01 |
|   8 |       INDEX UNIQUE SCAN         | PS_KK_SOURCE_HDR   |     1 |       |     1   (0)| 00:00:01 |
|   9 |     INDEX UNIQUE SCAN           | PS_PO_HDR          |     1 |       |     1   (0)| 00:00:01 |
| 10 |    TABLE ACCESS BY INDEX ROWID | PS_PO_HDR          |     1 |    37 |     2   (0)| 00:00:01 |One can see Elapsed time(ms)=2,483,336
Is it 2483336 in millisecond that means 41 minutes ?
If 41 minutes it's a long qurey . Can we do something to optimize that ?
Thank you.

Hi,
1) the amount of time it takes is not enough grounds for optimization. Only SQL that hurts the business by its slowness needs to be optimized (e.g.: if you tune a nightly batch job to run from 4 hours to 4 minutes no one will probably notice it)
2) if there is a valid business case for optimizing this statement, the first step would be posting the query's text + plan with rowsource stats
Best regards,
Nikolay

Oracle Load Testing - Metrics

Hi,
I'm new to Performance Testing.Which of the common metrics reading (in network,in database,in webserver,network,system etc.) would help me if i conduct a test to check the performance of the three tier application structure.I want to make sure that it is enough to handle the application.
Thanks,

For a good start, you could follow the methodology described here : http://www.oracle.com/technetwork/oem/grid-control/overview/wp-rapid-1.pdf
Then, you could monitor CPU/Mem/io for each server and depending on your environment, maybe to monitor closely the DB & the middle tier.
If it is Oracle DB & WLS, we do recommand to use Oracle Enteprise Manager to monitor the environment and then to gather metrics (for example AWR/ADDR report or JVM diagnostic) from OEM to Load Testing console.
JB

How to generate AWR report by using ttStats?

Hi all,are there any links or documents to show how to use ttStats toolkit to generate a report like Oracle AWR report?
If no links or document available, could you tell me how to use it?
In fact I want to know how to do a performance analysis for my application and perform some tune.
If any other toolkits are more suitable please let me know and of course the guide that how to use them is much welcome.
Thanks.

The ttStats executable can be found in the <TimesTen_home>/support directory. Before you can run the executable within a command prompt, you'll need to source the ttenv.* file in the <TimesTen_home>/bin directory.
The output of ttStats -h is as follows:
ttStats [-latchstats on|off] [-display-options] [-xml file]
[-statlevel level] <dsn or connstr>
ttStats -s[napshot] [-latchstats on|off] -xml <file> <dsn or connstr>
ttStats -report <xmlfile1> <xmlfile2> <-text|-html> <report_file>
Where parameters in <> are required, and in [] are optional.
-display-options:
This option only works in monitoring mode as
screen filtering. ttStats only displays on the
screen statistics with specified prefixes.
Available prefix options include:
-cg, -ckpt, -conn, -db, -grid, -lock, -log, -repl,
-stmt, -txn, -xla, -zz, where the naming of prefixes
conforms to naming of statistics in TimesTen
sys.systemstats table. If no display option
is specified, all statistics are displayed
by default. For example, run command
ttStats -cg -stmt dsn=TptbmData
to display only cache group and statement
related statistics.
-h[elp] Print help message and exit.
-interval <secs> Collect a set of statistics every <secs> seconds
(default=10).
Note: setting this lower can negatively impact
performance
-xml <file> Write statistics to <file> in XML format. All
statistics will be written to <file> no matter
whether they are displayed on screen or not.
For example ttStats -xml mon.xml dsn=TptbmData
and ttStats -xml mon.xml -cg dsn=TptbmData
write the same file even if the latter only
shows cache related statistics on the screen.
-snapshot One snapshot of all statistics are saved to disk
in the specified file format, then the program
exits. This option should be used together with
the -xml option.
-statlevel <level>
Set the SysStats level, this value of level only
applies to sys.systemstats table.
-report <xmlfile1> <xmlfile2> <-xml|-text|-html> <report_file>
Compare two snapshots, and generate an AWR like
report. At present, each XML file contains only
one snapshot. The difference is computed as the
absolute value of statistic value in <xmlfile2>
minus that in <xmlfile1>. The result can be
written in HTML and plain text formats. XML is
not supported as output format at present.
As an example, you can run the utility as the instance admin user by doing something like:
./ttStats -s -xml tmp.xml <DSN>
You can then open tmp.xml with a web browser and see statistics about the database.
You can also compare two previously captured snaps by doing something like:
./ttStats -report tmp1.xml tmp2.xml -html tmpReport.html

Create AWR Reports

Can it be possible to generate the AWR reports and ADDM reports for a specific time period from OEM GRID and mail it to specified mail id everyday.

user13549752 wrote:
Can it be possible to generate the AWR reports and ADDM reports for a specific time period from OEM GRID and mail it to specified mail id everyday.As I know OEM Grid Control can not convert AWR reports to daily jobs, but we can write a PL/SQL script to create AWR/ADDM reports and mail them:
http://www.gokhanatil.com/2011/07/create-awr-and-addm-reports-and-send.html
Regards
Gokhan

ADDM/AWR report

Just a simple question. How old AWR/ADDM report i can run in default on 10gR2
Thanks

HI OP,
the answer by nicolos is correct,
what aman post is , he suppose to reply in different thread but unfortunately he posted here.. nothing else..
you can go for 1 week duration to generate AWR report
Thanks

Generate AWR report

Hi All,
DB - 10.2.0.4.0
I ned to generate AWR report for 30 minute time intervel. For expample, if I need to generate AWR report for the timmig 03:00 PM to 03:30 PM, DO I need to create snapshot manually at 3 pm and 3:30 pm.
exec dbms_workload_repository.create_snapshot(); at 03:00 PM
exec dbms_workload_repository.create_snapshot(); at 03:30 PM
Also, is there any way, if the current time is 01:00 pm and can we create snapshot for 03:00PM and 03:30 PM?
hare krishna

Hi Hare,
http://www.dba-oracle.com/oracle10g_tuning/t_generating_awr_report.htm
The final AWR report can be built by using the PL/SQL API provided in the dbms_workload_repository package.
Two procedures that generate AWR reports are awr_report_text and awr_report_html . These procedures generate the AWR report for the specified snapshot range in TEXT or HTML formats, respectively. The following script shows one way of retrieving the AWR text report for the particular snapshot range:
SELECT
output
FROM
TABLE
(dbms_workload_repository.awr_report_text
(37933856,1,2900,2911 )
If you want complete working scripts to generate AWR reports, I have them in the code depot of my Oracle Tuning book:
http://www.rampant-books.com/t_oracle_tuning_book.htm
Hope this helps . . .
Donald K. Burleson
Oracle Press author

Performance Report in EM

Is it possible to produce a performance report of a database in Enterprise Manager over 1 month to 1 year ???

Hello,
You need to change retention peirod of awr/addm report from 7 days to 1 month or 1 year ; u can also create a baseline and compare with your future reports.
Regards

Error in generating ADDM Report(Oracle 11g 64 bit EE on linux RHEL 5)

I collected .dmp file from production using awrextr.sql and imported in our development side using awrload.sql .
I am able to generate awr snapshots report out of it without any trouble.
But When I try to generate addm report using addmrpti.sql I am facing following error(Please see output pasted below)
Specify the Report Name
~~~~~~~~~~~~~~~~~~~~~~~
The default report file name is addmrpt_1_7149_7156.txt. To use this name,
press <return> to continue, otherwise enter an alternative.
Enter value for report_name:
Using the report name addmrpt_1_7149_7156.txt
Running the ADDM analysis on the specified pair of snapshots ...
begin
ERROR at line 1:
ORA-13711: Some snapshots in the range [7149, 7156] are missing key statistics.
ORA-06512: at "SYS.DBMS_ADVISOR", line 201
ORA-06512: at line 27
Generating the ADDM report for this analysis ...
ERROR:
ORA-13608: The specified name NULL is invalid.
ORA-06512: at "SYS.PRVT_ADVISOR", line 3122
ORA-06512: at "SYS.DBMS_ADVISOR", line 585
ORA-06512: at line 1
End of Report
Report written to addmrpt_1_7149_7156.txt
SQL>
Any clue or help will be really helpful for us.

hello,
have a look at this'
ORA-13711:Some snapshots in the range [string, string] are missing key statistics.
Cause: Some AWR tables encountered errors while creating one or more
snapshots in the given range. The data present in one or more of these missing
tables is necessary to perform an ADDM analysis.
Action: Look in DBA_HIST_SNAP_ERROR to find what tables are missing in
the given snapshot range. Use the ERROR_NUMBER column in that view
together with the alert log to identify the reason for failure and take necessary action to
prevent such failures in the future. Try running ADDM on a different snapshot range
that does not include any incomplete snapshots.thanks and regards
VD
Edited by: Dixit on Aug 31, 2009 1:52 AM
Edited by: Dixit on Aug 31, 2009 1:53 AM

AWR & ADDR Report

Similar Messages

Maybe you are looking for