New book on 11g ASM published

There are a couple of books on the market already which cover 11g features.
As always many of them do not really cover a lot of the new features because everything is so new and the books were written under preasure. So the have the “COVERS 11g” on the cover but inside it is mostly 10g stuff.

Here is one which is a MUST READ for everybody who is interested how Automatic Storage Management (ASM) really works:
Oracle Automatic Storage Management  
What makes this book especially interesting is the fact that the authors are the Oracle developers of ASM themselves, Nitin Vengurlekar and  Richard Long, and Murali Vallath., the president of the RAC SIG,  plus the architect of the technology Bill Bridge who wrote the foreword for the book .

This book keeps its promises and covers the 10g and 11g features of ASM in detail.
And it is an easy reader with a sence of humor:

“The ACD is a log that allows ASM to make atomic changes to multiple data structures. It is similar to the redo logs used by the Oracle RDBMS. Each instance gests a 42 MB chunk of the ACD. (As Douglas Adams explained in Hitchiker’s Guide to the Galary, “42″ is the answer to the ultimate question of life, the Univese and Everything.) …

 This is the way of reading I like!!

Thank you guys, I love it

New RMAN book for 11g

There is a new book on RMAN by
Sam R. Alapati, Darl Kuhn, Arup Nanda.

It is titeled RMAN RECIPES and is published by APRESS.

 These authors have already proven their expert competency by publishing a lot of highly interesting white papers in OTN and other books in the past. (Just search for their names in OTN!).

Here are the details of the book:

ISBN10: 1-59059-851-2

ISBN13: 978-1-59059-851-1

670 pp.

Published on Aug 2007

eBook Price: $30.00

Price: $59.99

Here are the contents as announced by APRESS:

Reliably back up and recover your database using Oracles Recovery Manager.

Utilize New Oracle Database 11g features such as the Data Recovery Advisor.

Let Oracle Database manage your backup files via the Flash Recovery Area.

Automate backup and recovery tasks by writing scripts.

Troubleshoot RMAN problems and optimize RMAN performance.

Interface with a Media Management Layer to manage your backup tape library.

Recover from the loss of a control file, loss of an online redo log, and other (hopefully!) unusual situations.Looks interesting, I will order it definitively!

Brand new book on RMAN Backup & Recovery out!

Finally it has been published:
the ultimat RMAN book for Oracle 10g
by M. Hart & R. Freeman!

I have been waiting for the 10g version of the book since quite a while already.
It was one of the very good books for Oracle 9i and I can tell you that the new one is as good. http://www.mhprofessional.com/product.php?cat=112&isbn=0072263172

ISBN: 0072263172 / 9780072263176

What I like most about the book:
- It contains a number of workshop for advanced recovery scenarios and
points out all these RMAN functionalities which are much too less used and know like BLOCK RECOVERY, PROACTIVE CHECKING FOR CORRUPT BLOCKS and so on.
- it has good index
- it contains info about memory utilization of RMAN for tuning

The book also covers new 10g features, such as Flashback, Transportable Tablespaces Cross Platform …

It is a must for everybody who needs to know how RMAN works and how to utilize it to the max.

Table of contents
Part I: Getting Started with RMAN in Oracle Database 10g

Chapter 1: Oracle Database 10g Backup and Recovery Architecture Tour

Chapter 2: Introduction to the RMAN Architecture

Part II: Setup Principles and Practices

Chapter 3: RMAN Setup and Configuration

Chapter 4: Media Management Considerations Chapter

5: Oracle Secure Backup

Chapter 6: Enhancing RMAN with VERITAS NetBackup for Oracle Chapter

7: Configuring EMC NetWorker Module for Oracle Chapter

8: RMAN and Tivoli Storage Manager

Part III: Using RMAN Effectively Chapter

9: RMAN Backups Chapter

10: RMAN Restore and Recovery Chapter

11: Using Oracle Enterprise Manager for Backup and Recovery Chapter

12: RMAN Advanced Recovery Topics Chapter

13: Surviving User Errors: Flashback Technologies Chapter

14: Maintaining RMAN Chapter

15: Monitoring and Reporting on RMAN Chapter

16: Performance Tuning RMAN Backup and Recovery Operations

Part IV: RMAN in the Oracle Ecosystem

Chapter 17: Duplication: Cloning the Target Database

Chapter 18: RMAN and Data Guard

Chapter 19: RMAN and Real Application Clusters

Chapter 20: RMAN in Sync and Split Technology

Chapter 21: RMAN in the Workplace: Case Studies

Part V: Appendixes

Appendix A: RMAN Syntax Reference Guide

Appendix B: Exploring the Recovery Catalog

Appendix C: Setting Up an RMAN Test Environment

INDEX

connect lutz hartmann as sysdba

Oracle Database 10g Real Application Clusters Handbook

I have just received the recent new book (FEB. 2007!)
by K. Gopalakrishnan
Oracle Database 10g Real Application Clusters Handbook 

Gopalakrishnan is a Senior Principal Consultant in Oracle Corporation and was avarded “Oracle Author of the year 2005” by Oracle Magazine in for his 2005 published book “Oracle Wait Interface”, which was one of the few books of the last year about Oracle which were really worth their money. Pls see my post about this book also.

After having a first glance at the new book I am very enthusiastic about it, because it contains large chapters about cache fusion (36 pages!) and the Global Resource Directory GRD (>20 pages) explained from scratch , amoung other intereseting chapters about troubleshooting RAC, RAC deployment…
Again we find the high quality we already have enjoyed in Gopalakrishnan´s Wait Interface book, and I bet it will be a success story as well.

Table of contents

Part I: High Availability Architecture and Clusters
Chapter 1: Introduction to High Availability and Scalability
Chapter 2: Clustering Basics and History
Chapter 3: RAC Architecture
Part II: Installation, Configuration, and Storage
Chapter 4: RAC Preinstallation
Chapter 5: RAC Installation
Chapter 6: Automatic Storage Management
Part III: RAC Administration and Management
Chapter 7: RAC Basic Administration
Chapter 8: RAC Advanced Administration
Chapter 9: RAC Backup and Recovery
Chapter 10: RAC Performance Management
Part IV: Advanced Concepts in RAC
Chapter 11: Global Resource Directory
Chapter 12: A Closer Look at Cache Fusion
Chapter 13: Workload and Connection Management
Chapter 14: RAC Troubleshooting
Part V: Deploying RAC
Chapter 15: RAC Extensions
Chapter 16: Developing Applications for RAC
APPENDIX A: RAC REFERENCE
APPENDIX B: ADDING AND REMOVING CLUSTER NODES
APPENDIX C: REFERENCES
INDEX

ISBN 007146509X / 9780071465090

Paperback, 421 pages

connect lutz hartmann as sysdba

Active Session History in Oracle 10gR1+2

This article has been published in the Swiss Oracle User Group’s Newsletter 4/2006 (October 2006).

This article descibes the new on Oracle Database 10g Active Session History, a kind of data warehouse for session statistics.

A lot of the information in this post I have taken from the book
  Oracle Wait Interface by the
authors R.Shee, K.Deshpande and K.Gopalakrishnan
which I can strongly recommed to everybody
who has to deal with performance tuning for ORACLE of which release ever.
It covers the Oracle versions until 10gR1 and tells us everything we have ever wanted to know about latches, enqueues and all the wait events.

Starting with the version 10gR1 Oracle stores information about session activity on disk in the Automatic Workload Repository (AWR) which sits, amoung other things, in the new mandatory SYSAUX tablespace.
This information comes from statistics about active sessions which the Oracle Kernel collects automatically.
The active session history is enabled by default( influenced by the hidden parameter _ASH_ENABLE=true).

The main components of the ASH are
- two new background procresses, MMON and MMNL,
- a rolling buffer, which sits in the SGA fixed area and holds the historical statistical information of active sessions

This buffer has a minimum size of 1Mb in ORACLE 10gR1 and can grow to a maximum of either 30Mb, or 5% of SHARED_POOL_SIZE, or Numer of CPUs * 2Mb (whichever is smaller!).

This formula has been changed in 10gR2 to not linearly scale with
the number of CPUs.
All we get to know about the sizing is that “…the ASH will
auto-configure itself to try and hold at least one-hour worth of
instance activity without wasting too much value SGA space and will
never be bigger than 2% of SGA_TARGET or 5% of SHARED_POOL (if AutoSGA
is not used)“.

The contents of the rolling buffer get written down to disk on a regular basis by MMNL
( with every snapshot created and when the buffer gets full).
It gets flushed to the table WRH$_ACTIVE_SESSION_HISTORY in the AWR which is implemented like a facts table in a DWH with its columns as dimensions.
This is enbled with the hidden parameter _ASH_DISK_WRITE_ENABLE=true.

The information in the ASH is used, among others, by automatic tuning utilities like teh SQLTUNING Advisor (dbms_sqltune) for the Automatic Tuning Optimizer (ATO). This one will look for historical data about sql-statements and their exetutions in the ASH (see buttom) as well, anong other activities it can perform in order top find a potentially better execution plan.

There are a lot of DBA_ views we can use to read the historical information from the AWR:
SELECT table_name
FROM dictionary
WHERE table_name like ‘DBA/_HIST/_%’ ESCAPE ‘/’
ORDER BY table_name;

TABLE_NAME
—————————–
DBA_HIST_ACTIVE_SESS_HISTORY
DBA_HIST_BASELINE
DBA_HIST_BG_EVENT_SUMMARY
DBA_HIST_BUFFERED_QUEUES
DBA_HIST_BUFFERED_SUBSCRIBERS
DBA_HIST_BUFFER_POOL_STAT
DBA_HIST_COMP_IOSTAT
DBA_HIST_CR_BLOCK_SERVER
DBA_HIST_CURRENT_BLOCK_SERVER
DBA_HIST_DATABASE_INSTANCE
DBA_HIST_DATAFILE
DBA_HIST_DB_CACHE_ADVICE
DBA_HIST_DLM_MISC
DBA_HIST_ENQUEUE_STAT
DBA_HIST_EVENT_NAME
DBA_HIST_FILEMETRIC_HISTORY
DBA_HIST_FILESTATXS
DBA_HIST_INSTANCE_RECOVERY
DBA_HIST_INST_CACHE_TRANSFER
DBA_HIST_JAVA_POOL_ADVICE
DBA_HIST_LATCH
DBA_HIST_LATCH_CHILDREN
DBA_HIST_LATCH_MISSES_SUMMARY
DBA_HIST_LATCH_NAME
DBA_HIST_LATCH_PARENT
DBA_HIST_LIBRARYCACHE
DBA_HIST_LOG
DBA_HIST_METRIC_NAME
DBA_HIST_MTTR_TARGET_ADVICE
DBA_HIST_OPTIMIZER_ENV
DBA_HIST_OSSTAT
DBA_HIST_OSSTAT_NAME
DBA_HIST_PARAMETER
DBA_HIST_PARAMETER_NAME
DBA_HIST_PGASTAT
DBA_HIST_PGA_TARGET_ADVICE
DBA_HIST_PROCESS_MEM_SUMMARY
DBA_HIST_RESOURCE_LIMIT
DBA_HIST_ROWCACHE_SUMMARY
DBA_HIST_RULE_SET
DBA_HIST_SEG_STAT
DBA_HIST_SEG_STAT_OBJ
DBA_HIST_SERVICE_NAME
DBA_HIST_SERVICE_STAT
DBA_HIST_SERVICE_WAIT_CLASS
DBA_HIST_SESSMETRIC_HISTORY
DBA_HIST_SESS_TIME_STATS
DBA_HIST_SGA
DBA_HIST_SGASTAT
DBA_HIST_SGA_TARGET_ADVICE
DBA_HIST_SHARED_POOL_ADVICE
DBA_HIST_SNAPSHOT
DBA_HIST_SNAP_ERROR
DBA_HIST_SQLBIND
DBA_HIST_SQLSTAT
DBA_HIST_SQLTEXT
DBA_HIST_SQL_BIND_METADATA
DBA_HIST_SQL_PLAN
DBA_HIST_SQL_SUMMARY
DBA_HIST_SQL_WORKAREA_HSTGRM
DBA_HIST_STAT_NAME
DBA_HIST_STREAMS_APPLY_SUM
DBA_HIST_STREAMS_CAPTURE
DBA_HIST_STREAMS_POOL_ADVICE
DBA_HIST_SYSMETRIC_HISTORY
DBA_HIST_SYSMETRIC_SUMMARY
DBA_HIST_SYSSTAT
DBA_HIST_SYSTEM_EVENT
DBA_HIST_SYS_TIME_MODEL
DBA_HIST_TABLESPACE_STAT
DBA_HIST_TBSPC_SPACE_USAGE
DBA_HIST_TEMPFILE
DBA_HIST_TEMPSTATXS
DBA_HIST_THREAD
DBA_HIST_UNDOSTAT
DBA_HIST_WAITCLASSMET_HISTORY
DBA_HIST_WAITSTAT
DBA_HIST_WR_CONTROL
78 rows selected.

The rolling buffer can be viewed by querying the relevant v$view:
SELECT table_name
FROM dictionary
WHERE table_name like ‘V$%SESSION%HISTORY’
ORDER BY table_name;

TABLE_NAME
——————————
V$ACTIVE_SESSION_HISTORY
V$SESSION_WAIT_HISTORY
2 rows selected.
You can use a query like this to look at the session activity in a certain peride of time with a query like this:
SELECT session_id, event, COUNT(*), SUM(time_waited)
FROM v$active_session_history
WHERE session_state= ‘WAITING’
AND time_waited > 0
AND sample_time >= (sysdate-&MinutesAgo/(24*60))
GROUP BY session_id, event;

Enter value for minutesago: 5
old 5: AND sample_time >= (sysdate-&MinutesAgo/(24*60))
new 5: AND sample_time >= (sysdate-5/(24*60))

SESSION_ID EVENT COUNT(*) SUM(TIME_WAITED)
—————- ——– ————- ——————————
149 db file sequential read 3 40857
149 db file scattered read 1 1530
164 db file sequential read 1 12208

The autors also describe how to dump the contents of the rolling buffer to a trace file in USER_DUMP_DEST in order to use sqlldr to load the information into another database.

ALTER SESSION SET events ‘immediate trace name ashdump, level 10′;

You can also use the utility oradebug to dump the rolling buffer:

SYS @10gR2 SQL > oradebug setmypid
Statement processed
SYS @10gR2 SQL > oradebug unlimit
Statement processed
SYS @10gR2 SQL > oradebug dump ashdump 10
Statement processed
SYS @10gR2 SQL > oradebug tracefile_name
Statement processed
SYS@10gR2 SQL > oradebug tracefile_name
/u01/app/oracle/admin/orcl/udump/orcl_ora_5851.trc

– first lines: column names for data
– trace data displayed as comma separated list of values
– can use sqlldr to load the data into another DB for further analysis
– can be helpfull in cases of hanging system

connect lutz hartmann as sysdba