Skip navigation.

Feed aggregator

Next Generation Outline Extractor Webcast - Oct 8th!

Tim Tow - Fri, 2014-09-12 16:07
I am doing a repeat of my Kscope15 'Best Speaker' award-winning presentation as part of the ODTUG webcast series. Here is the official announcement from ODTUG:
Wednesday, October 8, 2014 12:00 PM - 1:00 PM EDT

Next Generation Essbase Outline Extractor Tips and Tricks
Tim Tow, Applied OLAP
The Next Generation Outline Extractor is the follow-up to the classic OlapUnderground Essbase Outline Extractor used by thousands of Essbase customers. This session, which was the highest-rated session at Kscope15 in Seattle, explores some of the new capabilities of the Next Generation Outline Extractor, including command line operations exported directly to a relational database. Attend this session to learn how to leverage this free utility in your company. Make you you sign up to join my on October 8th; you can register here!.
Categories: BI & Warehousing

Best of OTN - Week of September 7th

OTN TechBlog - Fri, 2014-09-12 13:43
Database Community - Laura Ramsey, OTN Database Community Manager

Here are a few round ups of session that will be at Oracle OpenWorld in just a few weeks!  

Database Developer Hands-On-Labs
Oracle Database Service sessions,  from good friend Chris Warticki - @cwarticki

OTN DBA/DEV Watercooler Blog- What do you know about the Oracle RAC Stack? It's a compelling approach to Virtualization. Read on to see what you need to know.

Java Community - Tori Wieldt, OTN Java Community ManagerLAST CHANCE: Use code DDD4 to save $75 off a JavaOne Discover Pass.

JCP Activities at JavaOne 2014 - Read this blog to learn what's happening with the JCP program, JCP.Next, Adopt-a-JSR, meet the 12th Annual JCP Award nominees and potential new JCP Executive Committee (EC) members.

Driverless Cars and Java - Paul Perrone is a regular fixture at JavaOne, and this year's conference is no exception. He will give a session called "Automated Vehicle Testing with Java." Read the full story, "Java Takes the Wheel.

Friday Funny: “Whats the object-oriented way to become wealthy?”
A: Inheritance
Systems Community  -

SPARC M7 - A good read about “the forthcoming SPARC M7, the biggest and baddest SPARC processor that either Sun Microsystems or Oracle has ever created.”

Interesting Q&A on the OTN Solaris Community:
"I've imported a service, using inetconv, into the smf repository with a typo. The service is now MASKED so I can't remove it using svcadm delcust. Can a service be un-MASKED ? - Turns out the behavior of "svccfg import" and "svccfg delete fmri" has changed in Oracle Solaris 11. Go to the conversation to learn more.

Oracle's SPARC T5-4 Server Processor Upgrade Animation  - This animation shows you how to upgrade a SPARC T5-4 server from a one-processor to a two-processor configuration and much more.

Get involved in community conversations on the following OTN channels...

OOW - Focus On Support and Services for Enterprise Manager

Chris Warticki - Fri, 2014-09-12 08:00
Focus On Support and Services for Enterprise Manager   Monday, Sep 29, 2014

Conference Sessions

Best Practices for Maintaining and Supporting Oracle Database
Balaji Bashyam, Vice President, Oracle
Roderick Manalac, Consulting Tech Advisor, Oracle
11:45 AM - 12:30 PM Moscone South - 310 CON8270 Best Practices for Maintaining and Supporting Oracle Enterprise Manager
Farouk Abushaban, Senior Principal Technical Analyst, Oracle
2:45 PM - 3:30 PM Intercontinental - Grand Ballroom C CON8567 Oracle WebLogic Server: Best Practices for Troubleshooting Performance Issues
Laurent Goldsztejn, Principal Engineer, Proactive Support, Oracle
4:00 PM - 4:45 PM Moscone South - 270 CON8307 Oracle Exadata: Maintenance and Support Best Practices
Christian Trieb, CDO, Paragon Data GmbH
Jaime Figueroa, Senior Principal Technical Support Engineer, Oracle
Bennett Fleisher, Customer Support Director, Oracle
4:00 PM - 4:45 PM Moscone South - 310 CON8259 Tuesday, Sep 30, 2014

Conference Sessions

Best Practices for Maintaining Oracle Fusion Middleware
Ken Vincent, Senior Principal Technical Support Engineer, Oracle
10:45 AM - 11:30 AM Moscone West - 3022 CON8285 Oracle Database 12c Upgrade: Tools and Best Practices from Oracle Support
Agrim Pandit, Principal Software Engineer, Oracle
5:00 PM - 5:45 PM Moscone South - 310 CON8236 Wednesday, Oct 01, 2014

Conference Sessions

Taming the Wild West with Oracle Database Options
Mike Brotherton, JP Morgan Chase
Ashok Pandya, Consulting Solutions Director, Oracle
4:45 PM - 5:30 PM Intercontinental - Union Square CON3910 Wednesday, Oct 01, 2014

Conference Sessions

Proactive Support Best Practices: Oracle E-Business Suite Payables and Payments
Stephen Horgan, Senior Principal Technical Support Engineer, Oracle
Andrew Lumpe, Senior Principal Support Engineer, Oracle
2:00 PM - 2:45 PM Moscone West - 3006 CON8479 Thursday, Oct 02, 2014

Conference Sessions

Real-World Oracle Maximum Availability Architecture with Oracle Engineered Systems
Bill Callahan, Director, Products and Technology, CCC Information Services
Jim Mckinstry, Consulting Practice Director, Oracle
9:30 AM - 10:15 AM Intercontinental - Grand Ballroom B CON2335 Oracle E-Business Suite Architecture Best Practices: Tips from CBS
John Basone, CBS
Greg Jerry, Director - Oracle Enterprise Architecture, Oracle
12:00 PM - 12:45 PM Marriott Marquis - Salon 4/5/6* CON3829 Best Practices for Maintaining Your Oracle RAC Cluster
William Burton, Consulting Member of Technical Staff, Oracle
Scott Jesse, Customer Support Director, RAC, Storage & RAC Assurance, Oracle
Bryan Vongray, Senior Principal Technical Support Engineer, Oracle
12:00 PM - 12:45 PM Moscone South - 310 CON8252 Optimizing Oracle Exadata with Oracle Support Services: A Client View from KPN
Eric Zonneveld, Ing., KPN NV
Jan Dijken, Principal Advanced Support Engineer, Oracle
1:15 PM - 2:00 PM Moscone South - 305 CON7054   My Oracle Support Monday Mix

Monday, Sep 29

Join us for a fun and relaxing happy hour at the annual My Oracle Support Monday Mix. This year’s gathering is Monday, September 29 from 6:00 to 8:00 p.m. at the ThirstyBear Brewing Company – just a 3 minute walk from Moscone Center. Admission is free for Premier Support customers with your Oracle OpenWorld badge. Visit our web site for more details: 6:00 PM - 8:00 PM ThirstyBear Brewing Company Oracle Support Stars Bar & Mini Briefing Center

Monday, Sep 29

Ask the stars of Oracle Support your toughest questions, learn about proactive support tools and advanced support offerings, and win a prize at one of our 10-minute mini-briefings where you are sure to leave with valuable tips and best practices based on our experience supporting Oracle customers around the globe. 9:45 AM - 6:00 PM Moscone West Exhibition Hall, 3461 and 3908

Tuesday, Sep 30

Ask the stars of Oracle Support your toughest questions, learn about proactive support tools and advanced support offerings, and win a prize at one of our 10-minute mini-briefings where you are sure to leave with valuable tips and best practices based on our experience supporting Oracle customers around the globe. 9:45 AM - 6:00 PM Moscone West Exhibition Hall, 3461 and 3908

Wednesday, Oct 01

Ask the stars of Oracle Support your toughest questions, learn about proactive support tools and advanced support offerings, and win a prize at one of our 10-minute mini-briefings where you are sure to leave with valuable tips and best practices based on our experience supporting Oracle customers around the globe. 9:45 AM - 3:45 PM Moscone West Exhibition Hall, 3461 and 3908

To secure a seat in a session, please use Schedule Builder to add to your Schedule.

Watch: 5 Best Practices for Launching Your Online Video Game

Pythian Group - Fri, 2014-09-12 07:24

Warner Chaves, Principal Consultant at Pythian, has had the privilege of working with several companies on their video game launches, and is best known for his work with the highly anticipated release of an action-adventure video game back in 2013. Through his experience, he’s developed a set of best practices for launching an online video game.

“You don’t want to have angry gamers on the launch of the game because they lost progress in the game,” he says. “Usually at launch, you will have really high peaks of volume, and there might be some pieces of the infrastructure that are not as prepared for that kind of load. There also might be some parts of the game that are actually more popular than what  you expected.”

Watch his latest video below, 5 Best Practices for Launching Your Online Video Game.

Categories: DBA Blogs

Log Buffer #388, A Carnival of the Vanities for DBAs

Pythian Group - Fri, 2014-09-12 07:22

In order to expand the knowledge about database features of any kind, blogs are inevitable these days. Whether its Oracle, MySQL, or SQL Server blogs writers are contributing like never before and this log buffer edition skims some of it.


The Oracle Utilities family of products use Oracle standard technology such as the Oracle Database and Oracle Fusion Middleware (a.k.a. Oracle WebLogic).

OBIEE SampleApp in The Cloud: Importing VirtualBox Machines to AWS EC2.

The default value for the INMEMORY_MAX_POPULATE_SERVERS parameter is derived from the PGA_AGGREGATE_LIMIT parameter.

Most customers of Oracle Enterprise Manager using JVM Diagnostics use the tool to monitor their Java Applications servers like Weblogic, Websphere, Tomcat, etc.

Taking Enterprise File Exchange to the Next Level with Oracle Managed File Transfer 12c.

SQL Server:

The concept of a synonym was introduced in SQL Server 2005. Synonyms are very simple database objects, but have the potential to save a lot of time and work if implemented with a little bit of thought.

This article summarizes the factors to consider and provide an overview of various options for HA and DR in cloud based SQL Server deployments.

Chris Date is famous for his writings on relational theory. Chris took on the role of communicating and teaching Codd’s relational theory, and reluctantly admits to a role in establishing SQL as the dominant relational language.

Introduction of how to design a star schema dimensional model for new BI developers.

Have you ever wondered why the transaction log file grows bigger and bigger? What caused it to happen? How do you control it? How does the recovery model of a database control the growing size of the transaction log? Read on to learn the answers.


A common migration path from standalone MySQL/Percona Server to a Percona XtraDB Cluster (PXC) environment involves some measure of time where one node in the new cluster has been configured as a slave of the production master that the cluster is slated to replace.

How to shrink the ibdata file by transporting tables with Trite.

OpenStack users shed light on Percona XtraDB Cluster deadlock issues.

There are a lot of tools that generate test data.  Many of them have complex XML scripts or GUI interfaces that let you identify characteristics about the data. For testing query performance and many other applications, however, a simple quick and dirty data generator which can be constructed at the MySQL command line is useful.

How to calculate the correct size of Percona XtraDB Cluster’s gcache.

Categories: DBA Blogs

Unusual Deadlock

Jonathan Lewis - Fri, 2014-09-12 06:08

Prompted by a question on OTN I came up with a strategy for producing an ORA-00060 deadlock that DIDN’T produce a deadlock graph (because there isn’t one) and didn’t get reported in the alert log (at least, not when tested on It’s a situation that shouldn’t arise in a production system because it’s doing the sorts of things that you shouldn’t do in a production system: but possibly if you’re trying to do some maintenance or upgrades while keeping the system live it could happen. Here’s the starting code:

drop procedure p2;
drop procedure p1;

drop table t1 purge;
create table t1 (n1 number);

insert into t1 values(1);
insert into t1 values(3);

create or replace procedure p1
        update t1 set n1 = 2 where n1 = 1;
        update t1 set n1 = 4 where n1 = 3;

create or replace procedure p2

        procedure q
                pragma autonomous_transaction;
                execute immediate 'drop procedure p1';


        update t1 set n1 = 4 where n1 = 3;


Of course you’re asking for trouble if you start doing DDL as part of your production code; and you’re asking for trouble if you start playing around with autonomous transaction; and if you use one to do the other you’re almost guaranteed to hit a problem eventually. All it takes in this case is a simple sequence of actions followed by a short wait (ca. 10 seconds):

Session 1: execute p1
Session 2: wait a couple of seconds, then execute p2

I got the following result from session 2:


ERROR at line 1:
ORA-00060: deadlock detected while waiting for resource
ORA-06512: at "TEST_USER.P2", line 8
ORA-06512: at "TEST_USER.P2", line 14
ORA-06512: at line 1

While this doesn’t show up in the alert log, I do get a trace file dumped for the session; in fact I got a trace file from both processes. The trace from the process that reported the deadlock started like this:


  Performing diagnostic dump and signaling ORA-00060

  Complete deadlock information is located in the trace file of process (pid: 8, osid: 17861, DIA0)
    trace file: /u01/app/oracle/diag/rdbms/test/test/trace/test_dia0_17861.trc


*** 2014-09-09 12:44:13.427

    dump requested by process (pid: 8, osid: 17861, DIA0)
    trace file: /u01/app/oracle/diag/rdbms/test/test/trace/test_dia0_17861.trc

The other process didn’t mention a deadlock, but started at the “HUNG PROCESS DIAGNOSTIC” line – the trace file had been triggered by the DIAG process. Of course, the DIAG process also dumped a trace file, summarising the situation; and this started with the following important note:

*** 2014-09-09 12:44:13.426

DEADLOCK DETECTED (id=0xf678cfe3)

Chain Signature: 'enq: TX - row lock contention'<='library cache pin' (cycle)
Chain Signature Hash: 0x39f9868d

The rest of the trace file told me what the other two processes had been doing when the ORA-00060 was initiated, but the point I want to pick up here is that we have a very brief summary in the “Chain Signature” that tells us we’ve had a collision between some DML (enq: TX) and some sort of cursor-like problem (library cache pin) and not a simple data cross-over.

If you’ve looked at the original OTN posting you’ll see that the Chain Signature in that case is “Chain Signature: ‘library cache lock'<=’library cache pin’ (cycle)”, which indicates a collision restricted entirely to the library cache (the lock suggests someone is using a package/cursor while the pin suggests that another session is trying to destroy/(re)compile it – and they’re each trying to do it to each other’s package ! (In the past when I’ve a deadlock of this type it’s been reported as ORA-04020 rather than ORA-00060.) I still have to work out exactly how the OP got into their deadlock (especially in view of their latest comment) -but since I don’t have the right version of Oracle to hand, and it might be a version-specific bug, I don’t think I’m going to try.


Oracle Database – Getting started with JSON Path Expressions

Marco Gralike - Fri, 2014-09-12 03:52
Yesterday my colleague Alex and I had the pleasure to do some extra presentations during AMIS’s Oracle OpenWorld preview evening. While still not getting around...

Read More

Retro Day - Workload Management in WebLogic Server

Steve Button - Thu, 2014-09-11 21:38
Quite recently, I stumbled across this older but still very relevant whitepaper on the workload management capabilities of  WebLogic Server.  Written by one of the key engineers at the time, who later went on to become an architectect, it explains the workings of the new work load management feature introduced into WebLogic Server at the time and covers the concepts of work-manager configuration and effect, work scheduling and prioritization, overload protection and  more.

This is well worth a read if you are ever looking for good information on the work load management feature of WebLogic Server:

Oracle OEM Cloud Control - AWR Warehouse

Yann Neuhaus - Thu, 2014-09-11 20:42

This post explains how to configure and use the new AWR warehouse functionality present in Enterprise Manager Cloud Control This new feature offers the possibility to store date from the AWR of your main databases in the EM AWR warehouse.

As OEM AWR warehouse automatically extracts the AWR data of the database you have selected, there is no impact on your production databases. The main advantage is that it allows to keep AWR historical data beyond the retention period of the target databases. Another benefit is a performance improvement and a space gain of the target databases because AWR data are uploaded in a centralized AWR warehouse.

We have to apply a patch to get the AWR warehouse functionality available in the menu Database Performance. For Linux x86-64, you have to apply the patch 19176910 which contains the EM DB plugin bundle patch Before applying of the patch the screen appears as follows:




We have to use the latest Opatchversion available for 11.1 release.

Watch out OPatch 11.2.0.x.0 version is not supported with Oracle Management Service (OMS) and


oracle@vmtestoraem12c:/u01/app/oracle/MiddleWare_12cR4/oms/OPatch/ [oms12c] opatch version

OPatch Version:

OPatch succeeded.


Then we can run:


oracle@vmtestoraem12c:/home/oracle/19176910/ [oms12c] opatchauto apply -analyze

OPatch Automation Tool

Copyright (c) 2014, Oracle Corporation. All rights reserved.

OPatchauto version :

OUI version       :

Running from       : /u01/app/oracle/MiddleWare_12cR4/oms

Log file location : /u01/app/oracle/MiddleWare_12cR4/oms/cfgtoollogs/opatch/opatch2014-08-25_16-45-18PM_1.log

OPatchauto log file: /u01/app/oracle/MiddleWare_12cR4/oms/cfgtoollogs/opatchauto/19176910/opatch_oms_2014-08-25_16-45-20PM_analyze.log

[Aug 25, 2014 4:46:33 PM]   Prerequisites analysis summary:


                             The following sub-patch(es) are applicable:







We can check if any patches are installed in the OMS:

oracle@vmtestoraem12c:/home/oracle/19176910/ [oms12c] opatchauto lspatches

OPatch Automation Tool

Copyright (c) 2014, Oracle Corporation. All rights reserved.

There are no patches installed in the OMS system.


Then we stop the OMS:


oracle@vmtestoraem12c:/home/oracle/19176910/ [oms12c] emctl stop oms

Oracle Enterprise Manager Cloud Control 12c Release 4

Copyright (c) 1996, 2014 Oracle Corporation. All rights reserved.

Stopping WebTier...

WebTier Successfully Stopped

Stopping Oracle Management Server...

Oracle Management Server Successfully Stopped

Oracle Management Server is Down


Then we run under the patch directory : opatchauto apply

Once the patch is applied you can start the oms with the classical command : emctl start oms

Finally we have the AWR warehouse menu:



We have first to configure the AWR warehouse repository:





You select the configure button:




You enter the database and host credentials:




You configure the snapshot management for a one year retention period, you can also define the snapshot upload interval from 1 hour to 24 hours:




The caw_load_setup job_%id is successful.

Now we have access to the AWR warehouse screen where we can add databases:




When you select add, you have to define preferred credentials to the database you add and add some grants:




We have to allocate the execute privilege on dbms_swrf_internal to system:


SYS> grant execute on sys.dbms_swrf_internal to system;

Grant succeeded.

For the NMO error, I forgot to run the agent 12c script:

[root@vmtestoradg1 ~]# cd /u00/app/oracle/agent12c/core/

[root@vmtestoradg1]# ./

Finished product-specific root actions.

/etc exist


Finally the databases are added:




We have to grant the access to the databases added, we select a target database and we choose Privileges:




We select the sysman administrator account:




As we have configured the snapshot upload interval to 24 hours, we will have data every day:




From this dashboard you can make ADDM comparaisons, view ADDM reports, ASH analytics or directly go to the performance page of the database you have selected.

The AWR warehouse feature requires the diagnostic pack license; This new feature seems very interesting due to the central and consolidated AWR warehouse. I will test this exciting feature deeper in the next weeks.

Oracle OpenWorld Sessions - Oracle Utilities

Anthony Shorten - Thu, 2014-09-11 20:04

I will be presenting three sessions at Oracle Openworld this year. Feel free to come to the sessions if you are attending and have a chat with me on your implementations and general feedback.

The three sessions I am conducting are:

I will also be around the Oracle Utilities area for general discussions with customers and also be appearing at demo grounds.

See you there.

Using Oracle's MAA with Oracle Utilities Products

Anthony Shorten - Thu, 2014-09-11 19:47

The Oracle Utilities family of products use Oracle standard technology such as the Oracle Database and Oracle Fusion Middleware (a.k.a. Oracle WebLogic). As we use those technologies the features and techniques for those products in Oracle's Maximum Availability Architecture can be applied to Oracle Utilities products.

Oracle Maximum Availability Architecture is a set of processes and guidelines for implementing high availability solutions. For Oracle Utilities products the following MAA solutions are applicable:

Other MAA documents are also applicable for Oracle Exadata, Oracle ExaLogic and Oracle VM. The guidelines and processes can be used to design and configure specific MAA solutions for Oracle Utilities products at your site.

Documentum upgrade project - Documentum Administrator and ACS

Yann Neuhaus - Thu, 2014-09-11 19:15

After having installed the Documentum Administrator, one of our test was to open a document. When I opened one for the first time the content was empty although I knew it contained text. But the correct content was well displayed when I opened it a second time.

In the ucf trace file I had:


FINE: pairs: {null: HTTP/1.1 500 Internal Server Error}{Server: Apache-Coyote/1.1}{Content-Type: text/html;charset=utf-8}{Content-Length: 3437}{Date: Mon, 27 Jan 2014 12:55:47 GMT}{Connection: close} Jan 27, 2014 1:55:47 PM com.documentum.ucf.client.logging.impl.UCFLogger error
SEVERE: Connection to ACS failed with response status : 0 Jan 27, 2014 1:55:47 PM com.documentum.ucf.client.logging.impl.UCFLogger debug
FINE: Failed processing segment Connection to the ACS failed - status : 0  at com.documentum.ucf.common.transport.impl.HTTPRemoteContentNativeConnector.checkConnectStatus(  at com.documentum.ucf.common.transport.impl.HTTPRemoteContentNativeConnector.prepareInputStream(
at  at Source)



To solve the issue, I had to replace in




EMC validated my workaround, logged an incident report, and reported a regression bug to be solved in future versions.

Full Disclosure

Floyd Teter - Thu, 2014-09-11 17:16
Over the next few weeks, I'll be sharing plenty of information relating to Oracle OpenWorld.  Mostly on this blog or via my Twitter account (@fteter).  As always, I try to share information in an accurate and a positive way.  I'm an Oracle fan, but I always try to balance that with accuracy and honesty.  In that spirit, it's important for me to make some disclosures about my attendance at OOW14.

I'm attending on "The King's Shilling".  Oracle generously covers travel costs for Oracle ACE Directors attending OOW and JavaOne, and I'm fortunate enough to be included in that community.  In addition, I've received a complimentary press pass to both OOW and JavaOne.  Yes, I am very appreciative - I freely admit that I'm a lucky guy in this regard.

Does Oracle's magnanimous funding of my OOW attendance buy them influence in regards to my opinions?  No.  And, so far as I can tell, Oracle has no expectations of influencing my opinions with their funding.  In fact, it's usually just the opposite - they've encouraged me to be honest in my opinions.  Frankly, I think they'd boot my big bottom out of the conference and ACE Director program if I were anything but honest in my opinions.

So I'll continue to be honest in sharing information relating to OOW14.  But, in the interests of transparency and fairness, I thought ya'all should be aware of the relationship.

Oracle Priority Support Infogram for 11-SEP-2014

Oracle Infogram - Thu, 2014-09-11 15:55

From Appsconsultants.comsome good general pointers for attendees: This year’s OpenWorld expect 70000 attendees, 10 tips to maximize your conference experience.
From Chris Warticki's Blog - Oracle Support: All Access Pass to Oracle Support.
Pythian is one of our favorite companies here at the Infogram. Here’s their rundown on their Oracle OpenWorld presence.
WebCenter customers, don’t miss this announcement from the Oracle WebCenter blog: Oracle WebCenter & Oracle BPM Customer Appreciation Reception.
From Fear and Loathing of Oracle COMMIT; Try ORA ROWSCN.
From Richard Foote’s Oracle Blog: Introduction to Attribute Clustering (The Division Bell).
From Oracle’s Building Bridges. Connecting realities blog this article: DBA in 5 minutes ! I don’t think you’ll actually want to go for the DBA interview five minutes after reading, but if you’re a sys admin or power user sometimes saddled with DBAish things when the DBA is unavailable this is a good one to print out and stash behind a piece of glass with a little hammer and a sign saying Break Glass to Start Database..Storage
Oracle, Parallelism and Direct Path Reads… on Flash, from flashdba.
Oracle Database: Script to Purge aud$ Table Using dbms_audit_mgmt Package, from Pythian.
SQL Developer
From that JEFF SMITH, a 30-part series. Here’s a taste: 30 SQL Developer Tips in 30 Days, Day 3: Syntax Highlighting.
From RittmanMead: Using Oracle GoldenGate for Trickle-Feeding RDBMS Transactions into Hive and HDFS
Virtualizationinfolet’s us know: Release: Oracle VM VirtualBox 4.3.16
From Fusion Applications Developer Relations: Integrating With Fusion Application Using Services (SoapUi - SSL).
From Oracle SOA Suite - Team Blog: Taking Enterprise File Exchange to the Next Level with Oracle Managed File Transfer 12c.
Java 9: A Global Community Is Behind Programming's Next Big Advance, from Forbes.
From OTN: Using the New C++11 Array and Tuple Containers.
UNIX on the Mac
Most Mac users don’t open up the tremendous potential of using their OS—Unix. Here’s a good article from The Terminal.
OpenStack 101 - How to get started on Oracle Solaris 11, from OpenStack on Oracle Solaris.
From the Oracle E-Business Suite Support Blog:
WEBCAST - Tools to Simplify your EBS Upgrade
Logistics and Inventory Consolidated RUP10 Patches Released
New Fixed Assets White Paper for Insurance Policies and Calculations
How to Fix FRM- and ORA- Errors in Oracle Teleservice
Invoice Validation Parallel Processing New Functionality Available
Changes to Service Request Logging - Guided Resolution Flows
Relational Theory Light
The lighter side of relational theory, from John Myles White: The Lesser Known Normal Forms of Database Design
…and Finally
Time to run this set of life hacks again:
35 Genius Life Hacks Everyone Should Know, Especially #18, from the San Francisco Globe.

Virtual Circuit Wait

Bobby Durrett's DBA Blog - Thu, 2014-09-11 15:28

On Monday we had some performance problems on a system that includes a database which uses shared servers.  The top wait was “virtual circuit wait”.  Here are the top 5 events for a 52 minute time frame:

Top 5 Timed Foreground Events

Event Waits Time(s) Avg wait (ms) % DB time Wait Class virtual circuit wait 1,388,199 17,917 13 50.98 Network db file sequential read 1,186,933 9,252 8 26.33 User I/O log file sync 1,185,620 6,429 5 18.29 Commit DB CPU 5,964 16.97 enq: TX – row lock contention 391 586 1499 1.67 Application

From other monitoring tools there was no sign of poor performance from the database end but virtual circuit wait is not normally the top wait during peak times.  Overall for the time period of this AWR report the shared servers didn’t seem busy:

Shared Servers Utilization Total Server Time (s) %Busy %Idle Incoming Net % Outgoing Net % 111,963 38.49 61.51 15.99 0.01

We have seen virtual circuit waits ever since we upgraded to 11g on this system so I wanted to learn more about what causes it.  These two Oracle support documents were the most helpful:

Troubleshooting: Virtual Circuit Waits (Doc ID 1415999.1)


Evidently when you return a cursor from a package and the cursor includes a sort step a shared server will be hung up in a virtual circuit wait state from the time the cursor is first fetched until the application closes the cursor.  Our application uses cursors in this way so it stands to reason that our virtual circuit wait times we saw in our AWR report represent the time it took for our web servers to fetch from the cursors and close them, at least for the cursors that included sort steps.  So, if our web servers slow down due to some other issue they could potentially take longer to fetch from and close the affected cursors and this could result in higher virtual circuit wait times.

Here is a zip of a test script I ran and its output: zip

I took the test case documented in bug 5689608 and added queries to v$session_wait to show the current session’s virtual circuit waits.

Here are the first steps of the test case:

create or replace package cursor_package as
cursor mycursor is select * from test order by object_name;
 open cursor_package.mycursor;
create or replace procedure test_case is
l_row TEST%rowtype;
if cursor_package.mycursor%isopen then
fetch cursor_package.mycursor into l_row;
end if;

These steps do the following:

  1. Create a test table
  2. Create a package with a cursor that includes an order by to force a sort
  3. Open the cursor
  4. Create a procedure to fetch the first row from the cursor

At this point I queried v$session_wait and found no waits:

SQL> select * from v$session_event
  2  where sid=
  3  (SELECT sid from v$session where audsid=USERENV('SESSIONID')) 
  4  event='virtual circuit wait';

no rows selected

The next step of the test case fetched the first row and then I queried and found the first wait:

SQL> exec test_case;

SQL> select * from v$session_event
  2  where sid=
  3  (SELECT sid from v$session where audsid=USERENV('SESSIONID')) 
  4  event='virtual circuit wait';

       SID EVENT                          TIME_WAITED
---------- --------------------------------------------------------
       783 virtual circuit wait           0

Note that time_waited is 0 which means the time was less than one hundredth of a second.  Next I made my sqlplus client sleep for five seconds using a host command and looked at the wait again:

SQL> host sleep 5

SQL> select * from v$session_event
  2  where sid=
  3  (SELECT sid from v$session where audsid=USERENV('SESSIONID')) 
  4  event='virtual circuit wait';

       SID EVENT                             TIME_WAITED
---------- --------------------------------------------------------
       783 virtual circuit wait              507

Total time is now 507 centiseconds = 5 seconds, same as the sleep time.  So, the time for the virtual circuit wait includes the time after the client does the first fetch, even if the client is idle.  Next I closed the cursor and slept another 5 seconds:

SQL> begin
  2   close cursor_package.mycursor;
  3  end;
  4  /

PL/SQL procedure successfully completed.

Elapsed: 00:00:00.01
SQL> host sleep 5

SQL> select * from v$session_event
  2  where sid=
  3  (SELECT sid from v$session where audsid=USERENV('SESSIONID')) 
  4  event='virtual circuit wait';

       SID EVENT                                 TIME_WAITED
---------- --------------------------------------------------------
       783 virtual circuit wait                  509

The time waited is still just about 5 seconds so the clock stops on the virtual circuit wait after the sqlplus script closes the cursor.  If the session was still waiting on virtual circuit wait after the close of the cursor the time would have been 10 seconds.

This was all new to me.  Even though we have plenty of shared servers to handle the active sessions we still see virtual circuit waits.  These waits correspond to time on the clients fetching from and closing cursors from called packages.  As a result, these wait times represent time outside of the database and not time spent within the database.  These waits tie up shared servers but as long as they are short enough and you have shared servers free they don’t represent a problem.

– Bobby

p.s. This is on hp-ux 11.31 ia64 Oracle



Categories: DBA Blogs

Pearson’s Efficacy Listening Tour

Michael Feldstein - Thu, 2014-09-11 14:06

Back around New Year, Michael wrote a post examining Pearson’s efficacy initiative and calling on the company to engage in active discussions with various communities within higher education about defining “efficacy” with educators rather than for educators. It turns out that post got a fair bit of attention within the company. It was circulated in a company-wide email from CEO John Fallon, and the blog post and all the comments were required reading for portions of the company leadership. After a series of discussions with the company, we, through our consulting company, have been hired by Pearson to facilitate a few of these conversations. We also asked for and received permission to blog about them. Since this is an exception to our rule that we don’t blog about our paid engagements, we want to tell you a little more about the engagement, our rationale for blogging about it, and the ground rules.

The project itself is fairly straightforward. We’re facilitating conversations with a few different groups of educators in different contexts. The focus of each conversation is how they define and measure educational effectiveness in their respective contexts. There will be some  discussion of Pearson’s efficacy efforts at a high level, but mainly for the purpose of trying to map what the educators are telling us about their practices to how Pearson is thinking about efficacy in the current iteration of their approach. After doing a few of these, we’ll bring together the participants along with other educators in a culminating event. At this meeting, the participants will hear a summary of the lessons learned from the earlier conversations, learn a bit more about Pearson’s efficacy work, and then break up into mixed discussion groups to provide more feedback on how to move the efficacy conversation forward and how Pearson’s own efforts can be improved to make them maximally useful to educators.

Since both e-Literate readers and Pearson seemed to get a lot of value from our original post on the topic, we believe there would be value in sharing some of the ongoing conversation here as well. So we asked for and received permission from Pearson to blog about it. Here are the ground rules:

  • We are not getting paid to blog and are under no obligation to blog.
  • Our blog posts do not require prior editorial review by Pearson.
  • Discussions with Pearson during the engagement are considered fair game for blogging unless they are explicitly flagged as otherwise.
  • On the other hand, we will ask for Pearson customers for approval prior to writing about their own campus initiatives (and, in fact, will extend that courtesy to all academic participants).

The main focus of these posts, like the engagement itself, is likely to be on how the notion of efficacy resonates (or doesn’t) with various academic communities in various contexts. Defining and measuring the effectiveness of educational experiences—when measurement is possible and sensible—is a subject with much broader application’s than Pearson’s product development, which is why we are making an exception to our blogging recusal policy for our consulting engagements and why we appreciate Pearson giving us a free hand to write about what we learn.

The post Pearson’s Efficacy Listening Tour appeared first on e-Literate.

All Access Pass to Oracle Support

Chris Warticki - Thu, 2014-09-11 12:00

Looking for tips, recommendations and resources to help you keep your Oracle applications and systems running at peak performance? Want to find out how to get more out of your Oracle Premier Support coverage?

More than 500 experts from across Services and Support will be on hand at Oracle OpenWorld to answer your questions and share best practices for adopting and optimizing Oracle technology.

  • Find out what Oracle experts know about the best tools, tips and resources for supporting and upgrading Oracle technology. Attend one of our “Best Practices” sessions.
  • Stop by the Oracle Support Stars Bar to talk with support experts. Open daily @ Moscone West, Exhibition hall 3461.
  • See Oracle support tools in action at one of our demos.
View the schedule of all of our Oracle Premier Support activities at Oracle OpenWorld for more information. Visit the Services and Support Oracle OpenWorld Website to discover how you can take advantage of all Oracle OpenWorld has to offer. See you there!

Why use CASE when NVL will do?

Tony Andrews - Thu, 2014-09-11 11:42
I've found that many developers are reluctant to use "new" features like CASE expressions and ANSI joins. (By new I mean: this millennium.) But now they have started to and they get carried away.  I have seen this several times recently:     CASE WHEN column1 IS NOT NULL THEN column1 ELSE column2 END Before they learned to use CASE I'm sure they would have written the much simpler:     NVL (Tony Andrews

check if using tcps part II

Laurent Schneider - Thu, 2014-09-11 11:31

in your current session, as written there, check sys_context('USERENV', 'NETWORK_PROTOCOL')

in another session, you could grab some hints out of the network service banner. Do the maths, when it is not-not using ssl, it probably is…

select sid,program,
  case when program not like 'ora___@% (P%)' then
  (select max(case
      then 'TCP'
when NETWORK_SERVICE_BANNER like '%Bequeath%' 
      then 'BEQUEATH'
      then 'IPC'
      then 'SDP'
      then 'Named pipe'
      then 'TCPS' end)
    where i.sid=s.sid) end protocol
  from v$session s;

---------- --------------- --------
       415 sqlplus(TNS V1- BEQUEATH
       396 sqlplus(TNS V1- IPC     
         6 Toad            TCP     
         9 Toad            TCPS    
         1 oracle(DIAG)            
       403 Toad            TCP