Blog Feed

Unable to Extend OEM 13.3 Domain to Add an Additional OMS

While trying to add an additional OMS to the current OEM 13.3 GCDomain using the provisioning console there’s an error in the Install homes step.  

The error contains the following message:

Caused by: oracle.sysman.oii.oiif.oiifb.OiifbEndIterateException:
oracle.sysman.oii.oiif.oiifb.OiifbEndIterateException: SEVERE:Values for the following variables could not
be obtained from the command line or response file(s):
MYORACLESUPPORT_USERNAME(MyOracleSupportUsername)
Silent install cannot continue.

This issue is related to Bug 28290139 and the workaround can be found in MOS 2432095.1 After performing the workaround the OMS addition job needs to be executed from scratch. This resolves the issue.  

Thanks,
Alfredo

Oracle Autonomous Data Warehouse – What is it?

Oracle Autonomous Data Warehouse (ADW) is one of the so-called DB as a Service (DBaaS) or probably we should call it Data Warehouse as a Service (DWaaS) with the difference that this service is completely automatic and autonomous.  

This service is comprised by an Oracle Database (of course) already pre-configured for analytics, data lakes and data warehouse workloads. This Oracle Database (PDB) has 18c features running on Exadata hardware. Oracle Exadata software is 18c with features like In-Memory delivered from the cell server.  

There’s a service console that helps to manage the ADW services.

On top of this, Machine Learning (ML) tools help with data analysis and data models.   Development tools like SQL Developer are useful to create objects, load data and more.  

Let’s now talk about tasks that happen automatically:  

Automatic statistics gathering during direct-path load operations

Automatic tuning

Automatic Partitioning

Automatic In-Memory

Automatic Indexing (soon)

Automatic Compression

Automatic Tablespace Management

Automated backups and patching  

ADW is also capable of repair itself. Machine Learning is used to detect anomalies and uses pattern recognition to determine if this problem is already in the problem knowledge-base. If is a known problem, it will apply the fix automatically.  

On the security side all the information is encrypted at rest. This means that backups and all data in the tablespaces is encrypted using TDE.

The connectivity between ADW and the rest of the world is secured be default. Oracle provides a wallet file that contains all the connectivity information required to login.

If you open this file you’ll notice that all connectivity is being done through SSL.

 (security=(ssl_server_cert_dn=… 

Provisioning is quite easy using the service console.   From the main dashboard you can click on “Create a data warehouse” button.

Select the Workload type between Autonomous Data Warehouse ADW or Autonomous Transaction Processing ATP loads.
 
 
 
 
Next you choose a compartment where the ADW database is going to live. For this example I’m going to choose ADWTest01 compartment. It is not recommended to use the root compartment.
 
 
 
 
Then we are going to set the Display and Database names. For this example, I’m choosing 1 OCPU and 1 TB of storage. Keep in mind that ADW is elastic, this means we can add more OCPUs and Storage on the fly without the need for downtime.
 
 
 
 
 
From the Administrator Credentials section we are going to set the ADMIN account password.
 
 
 
 
 
Next is license type. You can choose to use a license that you already own or subscribe to a new one.
 
 
 
 
Last but not least is the Tags section. You can add a tag to this ADW database to better manage it inside your tenancy.
 
 
 
 
Let’s now click Create Autonomous Database and wait few minutes for OCI to provision it.
 
 
 
 
The orange ADW icon will turn green once the ADW database is fully provisioned.
 
 
 
 
In my next post I’ll cover how to connect o the ADW database and how to load data to it.
 
Thanks,
Alfredo

Collaborate 2019 Presentations

Collaborate 2019 will take place April 7-11 in San Antonio this year.
This time I have 3 presentations accepted and ready to roll starting April 7th.
Simplifying EBS 12.2 ADOP


In this session I’ll talk about basic ADOP concepts, how to us them and real-life problems and best practice recommendations.
Listen to your Database. How to Understand an Oracle AWR Report.


This session is one of my favorites. In this quick tip I’ll show you the method I use to quickly identify a bottleneck in the DB system using an AWR report.
SMART Performance Monitoring. Exadata and OEM 13c.


I’ll talk about a product that has been set aside by Oracle for a while. Yes, that is OEM 13c.
This session will cover how to monitor Exadata systems using OEM 13c, how to use the plug-in features to find the system’s performance and to alert in case of issues.
Don’t forget to also visit us at our Viscosity booth. I’ll see you there in case you have questions.

Click on the Viscosity logo for all the company events.

 Viscosity Events

Thanks,
Alfredo

Configure Oracle Cloud Infrastructure’s VCN Network

Oracle Cloud Infrastructure requires a VCN (Virtual Cloud Network) to be created in order to connect to DB Systems (Bare Metal and VM). There are basically tow options to configure this VCN.
·       Public Subnet with Internet Gateway
·       Private Subnet
The Public Subnet with Internet Gateway will expose the network to the Internet. The Internet Gateway is a virtual router that provides a path for traffic between he VCN and the Internet. This option is mainly recommended for non-production workloads.


The Private Subnet cannot be reached from the Internet. You will need to configure a Dynamic routing Gateway (DRG) with either FastConnect or IPSec VPN on your on-premises site.



Once the VCN is created, you can associate the Bare Metal or the VM system to it. The navigation menu to create the VCN is Menu -> Networking -> Virtual Cloud Networks



Then click on Create Virtual Cloud Network and follow the wizard.


Once you create your VCN it will show as a (green) available icon.



More configuration information below:
Thanks,
Alfredo

Avoid Transaction Errors Due To Tablespace Shortage

How many times we have had a call from application teams regarding a session error due to a tablespace (either DATA, TEMP or UNDO) getting 100% full?
How many times we had ask for an estimate of space used by a process, session or transaction?
There are several options that can help us overcome this type of issues.
a)    If you are using Oracle Enterprise Manager (if you don’t, you should), make use of Corrective Actions to manage your data tablespaces.
b)    If your issue is more with tablespace quotas, TEMP or UNDO you may want to enable RESUMABLE session
RESUMABLE is an option which allows the session to go into a suspended state in such a way that it won’t fail/abort. The session cannot wait forever though, this is controlled by RESUMABLE_TIMEOUT initialization parameter.
In order to enable this option at the session level you have to execute below SQL statement:
SQL> ALTER SESSION ENABLE RESUMABLE;                                                                                                    
Alertlog file will record if the session entered into suspended mode.
Some examples are:
###########
statement in resumable session ‘User TEST1(32), Session 3, Instance 2’ was
suspended due to
ORA-01536: space quota exceeded for tablespace ‘USERS’

###########
statement in resumable session ‘User TEST1(32), Session 15, Instance 1’ was
suspended due to
ORA-01562: failed to extend rollback segment number 4

###########
statement in resumable session ‘User TEST1(32), Session 23, Instance 1’ was
suspended due to
ORA-01652: unable to extend temp segment by 32 in tablespace TEMP

###########
statement in resumable session ‘User TEST1(54), Session 2, Instance 1’ was
suspended due to
ORA-01653: unable to extend table TEST1.TEST_RESUMABLE by 256 in tablespace USERS

While the session is into this suspended state it will account time to the “statement suspended, wait error to be cleared” wait event.


Once the error that sent the session to the suspended state is fixed, the session will resume its work.
More details on below Oracle’s documentation:
Thanks,
Alfredo

Create a Linux VM on Azure

In this post I’ll show you how easy is to create a Linux VM in Microsoft’s Azure cloud.
Once you login to Azure using your account you will be presented with the main dashboard.
Click on Virtual Machines and then click Add.
The first section contains the basic information of your VM.
Select a subscription. In my case is the Free Trial subscription.
For Resource group we are going to click Create new. A resource group is a logical way to organize your resources.
I just created an oracle-vm-resource-group.



Type a VM name and select the desired Region for your VM. The region is where the datacenter is located. You may want to choose the closest one to you. For availability I’m going to choose no infrastructure redundancy required as this is a test VM. There are different availability types for your VM. Below URL links to a document that explains the Azure regions and sets in detail.
The size of the VM will have a direct impact on your bill. Be careful when you chose the sizing as you don’t want to waste resources and money and on the other side you don’t want to have a system that is underperforming. Click on Change size and below table will show you all your options.

I selected 2vcpus, 8 GB memory VM.
For the ADMINISTRATOR ACCOUNT I’m going to chose SSH public key to connect to my VM.
In the SSH public key, I’m going to provide the public key of my local computer so I can gain access to the VM.
I don’t have a corporate Active Directory in my lab, so Login with Azure Active Directory is turned off.
Public inbound ports will allow SSH to connect.


Nor for Disks. There are 3 options. Premium SSD is the fastest option available and is used for production like workloads. Standard SSD still delivers good performance while the Standard HDD is the cheapest and slowest option.
I’m going to create a 300 GB Standard SSD disk for this VM.


For the NETWORK INTERFACE section my Virtual Network is going to be under my oracle-vm-resource-group. The Subnet and the Public IP will be created (new) and as previously selected the SSH inbound port is going to be open. 



For Management and Guest config I’ll leave the defaults and I’m going to jump to the Review + Create section. In this Review + create section you can see an estimate of your VM pricing.


Now, if you need to create multiple VMs like this, you may want to automate the process using the Azure CLI. Click on the Download a template for automation link located in the bottom right part of the page. I’ll cover Azure CLI in a later post.


Now click on Create and wait for your VM to be provisioned. You’ll receive a notification once the VM is created. 


Now is time to connect to it. From the VM dashboard find the Public IP address. Or you can click on connect button for more details.


Now connect to it using SSH. As we already shared our public key this should work without using a password.
$ ssh alfredokrieg@23.99.209.18
The authenticity of host ‘23.99.209.18 (23.99.209.18)’ can’t be established.
ECDSA key fingerprint is SHA256:yfdef8ToOBoI5a/FCogDbVMGCsdsZXmSylnH4yZ4sZs.
Are you sure you want to continue connecting (yes/no)? yes
Warning: Permanently added ‘23.99.209.18’ (ECDSA) to the list of known hosts.
$ hostname
oraclevm
Thanks,
Alfredo

Oracle Cloud Infrastructure – Database Service – Bare Metal

The Database service part of the Oracle Cloud Infrastructure offers the autonomous and user-managed database options. The user-managed can be customized and you can choose between using Bare Metal, Virtual Machine or Exadata systems. These are called DB Systems.
The autonomous database cannot be customized as they come pre-configured. You can choose between Autonomous Transaction Processing or Autonomous Data Warehouse.
Licensing wise there are two models. License included covers the costs of the license in the cloud service. Bring Your Own License (BYOL) allows to use costumer’s Unlimited License Agreement or Non-Unlimited License Agreement with Oracle Cloud Infrastructure. There are some restrictions that apply to BYOL. More details about pricing can be found in below link:
There are two ways to access Oracle Cloud Infrastructure. You can use the Console (browser) and the REST API. Below is the Oracle Cloud Infrastructure menu.
Let’s start looking at Bare Metal DB Systems. They are available as a single metal server running Oracle Linux 6.8 and NVMe storage locally attached. NVMe provides additional boost in performance for these systems.
When a bare metal DB system is created you select the Oracle Database Edition for the whole system. This means that you can have multiple DB homes with different versions but all with the same edition.
Bare metal comes in two shapes available. BM.DenselO2.52 that provides 1-node with up to 52 CPU cores, 768 GB of memory and eight 6.4 TB NVMe storage drives. BM.DenselO1.36 that is 1-node with up to 36 CPU cores, 512 GB of memory and nine 3.2 TB drives.




Database versions available are 11.2.0.4, 12.1.0.2, 12.2.0.1 and 18.0.0.0.

As you can see from the previous screenshot. You can opt for automatic backups going to an Oracle Cloud Object Storage. 

I’ll cover Virtual Machine and Exadata DB systems including backups in my next post. Stay tuned!
Thanks,
Alfredo