5.26.2020

Oracle Cloud & Third party tools

There have been several buzz words and offerings since the invention of Cloud concepts. We have been hearing of Private Cloud, Public Cloud, Hybrid cloud and more recently multi-clouds concepts and offering from many leading cloud market players.  All most all leading market players are offering several tools and solutions to support seamless administration, maintenance and operations tasks.

Despite many OCI offerings, still we sometimes might need third-party tools or solutions to achieve some of the goals to move or manage the cloud platforms. So, I gonna walk through some of the third-party tools/solutions that are certified and supported OCI operations.

  • According to Commvault, over 80% of companies are using the multi-cloud today. Commvault Cloud Backup solution provides support backup and recovery for  over 40% clouds. This can ease backup and recovery operations between clouds.
  • Vertitas's Netbackup 7.7.x supports backup and recovery operations on Oracle Cloud as well as supports multi-cloud data protection solutions. With NetBackup multi-cloud data protection, you can achieve seamless deduplication between data centers and multi-clouds.  
  • Using Cyberduck version 6.4.0 or higher, you can connect to OCI object storage
  • With dbm cloud systems, you can easy move data and VMs from Classic to OCI, migrate & replicate data from on-premise to OCI and move data stored from any cloud vendor into OCI.
  • VictorOps with OCI integration can greatly assist in cloud infrastructure and application monitoring, as well as real-time incident response will improve significantly. 
  • ZeroDown software provides HA option for Oracle cloud infrastructure. 
  • Coriolis is the simplest way to migrate your windows or Linux VMs alongside their underlying storage and networking configuration across cloud platforms.

There are many other third-party useful solutions available in the market. This is a just tiny list of useful third-party solutions on OCI.

5.14.2020

Migration methods to Autonomous Database (ADB) - Part IV (DBMS_CLOUD.COPY_DATA)

In this part of Migration methods to Autonomous Database (ADB) article series, we will see how to load data into ADB database using the DBMS_CLOUD package.

Assuming you have an ADB database up & running and you wanna load data from the file stored in any of the supported object storage. The image below from Oracle presentation depicts explains what type of object storage is supported to store the files, and settings required part of DBMS_CLOUD package to load the data.



Below image depicts the important procedures and table details to monitor the loading jobs:


You may review my earlier articles to understand how to load files into an Oracle object storage. Once you have the files put in a bucket, you then need to store the OSS credentials in the ADB database using the example below:



And finally use the example below to load the data into the table:

BEGIN
DBMS_CLOUD.COPY_DATA(
table_name =>'CHANNELS',
credential_name =>DEF_CRED_NAME,
file_uri_list =>'https://swiftobjectstorage.<region>.oraclecloud.com/v1/<tenancy>/<bucketname>/<filename>',
format => json_object('delimiter' value ',')
);
END;
/


If you are loading bulk data, you can query table below to know data loading operation progress:

SELECT * 
FROM  user_load_operations
WHERE type = 'COPY';
SELECT table_name, owner_name, type, status, logfile_table, badfile_table
FROM user_load_operations WHERE type = 'COPY';
Tables below provide information about log file and bad file of data load operations:

SELECT * FROM copy$1_log;
SELECT * FROM copy$1_bad;

So, we have learned how to load data into ADB using the DBMS_CLOUD package.


5.13.2020

Migration methods to Autonomous Database (ADB) - Part III (SQL Developer)

In part III article series of Migration methods to Autonomous Database we will see how to migrate or load data into ADB using SQL Developer.

Below slide from Oracle presentation depicts SQL Developer different use cases to migrate or load data into ADB:





Following screenshots walk through data loading scenario from an excel sheet into an existing table using SQL Developer. To walk though this demonstration, ensure the following is in-place:

  • an ADB is created and running
  • You have downloaded the database connection credentials (as shown in the below screenshots)
  • You have an excel sheet to load the data


Downloading connection credentials

Presuming an ADB is created and it's up & running. From the OCI console, select the ADB, click the DB Connection tab to download the credentails.







You have to input a password before you download the connection credentials.



Start the SQL Developer tool and connect to the ADB using the credentials downloaded previously.

Specify the location of the .zip files (connection credentials) and choose the suitable TNS service, high, medium or low.



Once you connected to the database, select table and choose import data option by clicking the right mouse button on the table, as shown below:

















In this scenario, there is a table 'EMP, and we will load the data from an excel sheet. So, choose the file. Ensure you select Local File option from the Source.







Choose Insert option from Import Method:



Map the columns, if not already selected.



Choose the Match by option drop down list:



Finish gives the import details.



Once you click the Finish button, data will be loaded successfully and below message will appear:



And you can see the data is successfully loaded and can be viewed.














So using SQL Developer, you can easily load the data from files CSV, XLS, XLSX, TSV and TXT.
Hope you got the basic understanding of data loading using SQL Developer.

5.12.2020

Migration methods to Autonomous Database (ADB) - Part II (MV2ADB)

In part II of migration methods to ADB blog series, we will look into MV2ADB tool details and usage. If you haven't read the Part I, I would encourage you to read it to get basic knowledge on various solutions offered by Oracle and some of the prerequisites for on-premise database migration to ADB.

Move to Autonomous Database (aka MV2ADB) tool new tool introduced by Oracle to easily migrate on-premise Oracle database over Autonomous Database levering Oracle data pump capabilities. The tool automates data loading into Oracle Object storage and migrating the data into ADB by one-click option.

The picture below depicts the flow of data migration process using the MV2ADB tool:



Ensure following conditions are met before you launch the migration operation:

  • HTTP connectivity between on-premise and Oracle object storage.  Required to move dump files
  • Download ADB credentials file
  • Download and install latest Oracle client to have new version of expdp/impdp and other essential tools
  • Perl >=5.10
  • Perl-data-dumper
  • Set java in execution path
  • Also download and configure OCI CLI from the below link:

https://github.com/oracle/oci-cli




The MV2ADB tool can be installed on-premise using the RPM shown below:

# rpm -i mv2adb-2.0.1-X.noarch.rpm

Following is the directory structure of the tool:

# tree /opt/mv2adb/
/opt/mv2adb/
├── conf
|   └── DBNAME.mv2atp.cfg
├── lib
|   ├── MV2ADB_LoggingAndTracing.pm
|   ├── MV2ADB_passwd.jar
|   ├── MV2ADB_PFile.pm
|   ├── MV2ADB_Queries.pm
|   └── MV2ADB_Utils.pm
└── mv2adb
└── utils
    ├── install_adb_advisor.sql
    └── premigration.jar


 You can also download the tool zip file mv2adb-2.0.1-X.tar.gz from the MOS specified here and using below tar command to untar the file into a specific directory on the server.

tar xvfz mv2adb-2.0.1.X.tar.gz -C <target directory>

Before you use the tool, ensue the path is set correct, as shown below:

$ export MV2ADB_HOME=/home/oracle/mv2adb/mv2adb-2.0.1.x


Using command below, you can de-install the tool:

# rpm -e mv2adb-2.0.1-X.noarch


You can refer the log files for diagnose issues related to the tool:

"/opt/mv2adb/out/log"


MV2ADB supports following operational modes:

  • Auto Operations
  • ExpDP Operations
  • ImpDP Operation
  • OCI Object Storage/Bucket Operations
  • Database Schema Operations
  • Encrypt Password Operations
The Auto operations (aka one command operation) automates the entire migration operation, performing the following individual operation:
  • Schema based expdp from source database
  • Upload the dump file over OCI object storage
  • Loading data with impdp to ADB
Refer the below sytanx for parameters and usage:


Courtesy MOS 2463574.1

Example with configuration file:
# ./mv2adb auto -conf conf/mv2atp.cfg

As mentioned the different operation modes, you can also use to export, copy, or import data using individual operation modes available with the tool. Refer the examples below:

# ./mv2adb expdp \
--dbcs //<host name/ip address>/DB122H1 \
--schemas SCHEMA_1,SCHEMA_2,SCHEMA_3,SCHEMA_4 \
--dumpname expdp.dmp --dumppath /tmp \
--ohome /u01/app/oracle/product/12.2.0.1/dbhome_1 \
--adbname RCATP \
--adbcfile /opt/mv2adb/source/Wallet_RCATP.zip \

# ./mv2adb impdp \
-ociregion us-phoenix \
-ocinamespace NAMESPACE_NAME \
-ocibucket BUCKET_NAME \
-ociid mail@corp.com \
-adbname RCATP \
-cfile /opt/mv2adb/source/Wallet_RCATP.zip \
-ichome /u01/app/oracle/product/instantclient_18_3 \
-dumpfile rcatp_exp_01.dmp,rcatp_exp_02.dmp \
-encryption \
-enctype AES256

For more usage Syntax and examples, refer MOS 2463574.1.

One of the advantages of MV2ADB is that the tool automatically install, configure and execute the Schema Advisor.

The tool can be used with configuration file or command line options.


References:

5.10.2020

Migration methods to Autonomous Database (ADB) - Part I (Data Pump)

In this part of on-premise Oracle database migration to ADB series, we will focus on set of tools to migrate your on-prime database to Autonomies database (ADB). We will briefly discuss about various migration solutions offered by Oracle and walk through data pumps migration pre-requisites and procedure.

Database cloud migration requires a good amount of knowledge about various methods offered by Oracle and the best solution that suits your business need. The best place to start with understanding various database cloud migration solutions is to visit the below interactive Cloud Migration Advisor page designed by Oracle:


ADB Schema Advisor

Once adequate knowledge is gained, the very next step is to know the restrictions & limitations imposed by Oracle ADB. Oracle simplifies this task by introducing a advisor, ADB Schema Advisor. Configure Oracle Autonomous Database Schema Advisor to analyze existing on-premise database and get a report on any concerns that may arise during migration phase. ADB Schema Advisor a very light weight utility with capabilities to analyze existing on-premise database for migration suitability. You can also easily drop the schema. As there are certain limitations with some data types in ADB, the utility helps analyzing and reporting such objects with restricted data types in on-premise database as part of analysis. The utility runs on existing schemas and produces following report:
  • Migration summary report with object count
  • List objects that can't be migrated to ADB due to restrictions and limitations
  • Also, objects list that can be migrated with modifications automatically during import process 
  • And Best practices and guidance are listed under Information section
You can download the ADB Schema Advisor utility from MOS 2462677.1, which also includes installation & executions steps and advisor output.

Migration methods

Below picture depicts various set of tools supported by Oracle to migrate your on-premise Oracle database to Autonomous Database:





Cloud migration path

To be able to migrate the data, dump files must be placed in cloud storage, either from Oracle, Microsoft Azure, AWS S3 or Oracle cloud object storage. 


Migrating with Oracle Data Pump

Oracle Data Pump offers flexible and fast data movement between Oracle databases. Data pump lets you import from dump files residing on cloud storage. Refer the following example to export on-premise database:

expdp sh/welcome123@mydb \
exclude=index,cluster,indextype,materialized_view,materialized_view_log,materialized_zonemap,db_link \
data_options=group_partition_table_data  \
parallel=16 \
schemas=sh \
dumpfile=exp_MYDB_%u.dmp \
encryption_pwd_prompt=yes
Below are some guideline and recommend parameters to use with expdp:

  • Schema level export for ADB migration
  • Use parallelism to speed-up export job
  • Exclude data types that are not supported in ADB with exclude parameter 
  • With data_options parameter, you can group portions into a non-partition table, recommended by ADB
Once the database is exported, move the dump files to Cloud Storage. In this blog, we will presume you have Oracle Cloud Object Storage option to put your dump files.

Download the latest Oracle client software on your desktop (if not exist) with also includes the latest Data Pump utilities. Before you start importing the database into ADB, ensure the following are in-place:

  1. An ADB database is configured, up and running
  2. You have downloaded the ADB connect credentials from OCI page for the ADB
  3. Export files are moved to Cloud storage
  4. Create cloud storage credentials in ADB 
For more details about credentials download, refer my earlier blog post.

If you are using Oracle cloud object storage, you will have to create a bucket and place all your dump files.

Create cloud storage credentials in ADB referring below example:

BEGIN
  DBMS_CLOUD.CREATE_CREDENTIAL(
    credential_name => 'DEF_CRED_NAME',
    username => 'adwc_user@example.com',
    password => 'Welcome12$'
  );
END;
/

Once the credentials are created in ADB, run the imp command from your desktop referring the example below:

impdp admin/password@ADWC1_high \       
     directory=data_pump_dir \       
     credential=def_cred_name \      
     dumpfile= https://objectstorage.us-ashburn-1.oraclecloud.com/n/namespace-string/b/bucketname/o/export%u.dmp \
     parallel=16 \
     encryption_pwd_prompt=yes \
     partition_options=merge \ 
     transform=segment_attributes:n \
     transform=dwcs_cvt_iots:y transform=constraint_use_default_index:y \
     exclude=index,cluster,indextype,materialized_view,materialized_view_log,materialized_zonemap,db_link
If you have worked with ADB, you know that admin is the default administrator user. If you wanna use different user to import, then, you will have to create that user in the ADB with required privileges for import. You can choose the TNS service name from the credentials file you have downloaded. You must specify the object URL in the dump file parameter, which is the location of your bucket with dump files.

You may notice we didn't mention the log file in the impdp. To review the log file, you may have to move the file to your object storage using the syntax below:

BEGIN
  DBMS_CLOUD.PUT_OBJECT(
    credential_name => 'DEF_CRED_NAME',
    object_uri => 'https://objectstorage.us-ashburn-1.oraclecloud.com/n/namespace-string/b/bucketname/o/import.log',
    directory_name  => 'DATA_PUMP_DIR',
    file_name => 'import.log');
END;
/

Conclusion
In nutshell, we have learned about various set of tools supported by Oracle to migrate on-prime database to ADB and a walk through a procedure of migrate an on-premise database to ADB using data pumps with best practices for fast migrations. In the subsequent blog posts, we will learn about MV2ADB and other tools and their procedure to migrate to ADB. 

References:

Refer some of below notes and links for more understanding about on-premise database migration to Oracle ADB.

Oracle Autonomous Database Schema Advisor (Doc ID 2462677.1)
https://docs.oracle.com/en/cloud/paas/autonomous-data-warehouse-cloud/user/load-data-data-pump.html#GUID-30DB1EEA-DB45-49EA-9E97-DF49A9968E24
https://www.oracle.com/middleeast/database/technologies/cloud-migration.html
https://www.oracle.com/webfolder/s/assets/webtool/cloud-migration-advisor/index.html

Note : Pictures & Diagrams copied from various Oracle presentations for eduction and demo purpose only.

4.04.2020

DB Links and ADB - Dos and Don'ts

DB Link simplifies data sharing across oracle databases. And DB link is created using the CREATE DATABASE LINK SQL command, a very simple and straight forward method. DB link creation has a different procedure on an Autonomous database in contract to typical Oracle databases. In this post, we will highlight the rules, limitations and the use of new DBMS_CLOUD_ADMIN package. Let's get into the action.

Rules and Limitations

Note some of below rules and limitations of creating DB link from ADB:

  • The target databases (non-ADB) should be configured to use TCP/IP with SSL (TCPS) authentication
  • The allowed DB port range is 1521-1525 to ensure the security. You can define the port while creating the DB link
  • ADB DB link support databases that are only accessible through a public IP or public hostname
  • One directory restricted to one wallet file. Create multiple directories to keep multiple wallet files of multiple databases
  • Supported target database versions are 11.2.0.2, 12.1.0.2, 12.2.0.1, 18c and 19c.
Creating DB link from Autonomous Database

Now let's walk through DB link creation procedure from an Autonomous database. As mentioned, you should use DBMS_CLOUD_ADMIN.CREATE_ DATABASE_LINK procedure on ADB. Below steps describe the required procedure:
  • Copy your target database wallet file (cwallet.sso) which contains target database certification to an object store
  • Using DBMS_CLOUD.CREATE_CREDENTIAL procedure, create access credentials to the object store where the target wallet file is stored.
  • Upload the target database wallet file from Object store to database directory using the DBMS_CLOUD.GET_OBJECT procedure. Refer example below:
BEGIN 
     DBMS_CLOUD.GET_OBJECT(
        credential_name => 'DEFF_CRED_CRDB',
        object_uri => 'https://objectstorage.us-phoenix-1.oraclecloud.com/n/namespace-string/b/bucketname/o/cwallet.sso',
        directory_name => 'DATA_PUMP_DIR'); -- DATA_PUMP_DIR is the predefined/default directory. It can be another DB directory.
END;
/
  • Create credentials on ADB, refer below example:

BEGIN
  DBMS_CLOUD.CREATE_CREDENTIAL(
    credential_name => 'DBL_CRDB',
    username => 'SCOTT', -- username must be specified in CAPITAL
    password => 'welcome1'
  );
END;
/
Details are stored in the database in encrypted format.
You can use the query below to list the stored credentials:

SELECT credential_name, username, comments FROM all_credentials;
  • You can use below command to drop credentials from the database:
BEGIN
   DBMS_CLOUD.DROP_CREDENTIAL('DBL_CRDB');
END;

    • Now we are ready to create database link from Autonomous database. Use the example below to create:
    BEGIN
         DBMS_CLOUD_ADMIN.CREATE_DATABASE_LINK(
              db_link_name => 'ADB_TO_CRDB', 
              hostname => 'adb.eu-frankfurt-1.oraclecloud.com', 
              port => '1525',
              service_name => 'example_medium.adwc.example.oraclecloud.com',
              ssl_server_cert_dn => 'CN=adwc.example.oraclecloud.com,OU=Oracle BMCS FRANKFURT,O=Oracle Corporation,L=Redwood City,ST=California,C=US',
              credential_name => 'DBL_CRDB',
              directory_name => 'DATA_PUMP_DIR');
    END;
    /
    If you are not an admin privileged user, get the permission to run DBMS_CLOUD_ADMIN.CREATE_DATABASE_LINK procedure.

    Now, you can use the DB link from ADB to the target database.

    To drop DB link from ADB, you can use the example below:

    BEGIN
         DBMS_CLOUD_ADMIN.DROP_DATABASE_LINK(
              db_link_name => 'ADB_TO_CRDB' );
    END;
    /

    Creating DB link to ADB

    We have learned so far how to create a DB link from ADB to target database. This section will zip through DB link creation to ADB.
    As I have mentioned my previous blog entry that ADB only accepts secured client connections. To which, you will have to download client connect credentials. You can refer my previous blog post for the procedure.

    Before creating database link, ensure the following is met:
    • Download the client connect credentials
    • Unzip the downloaded file to a secure location your computer
    • Turn of GLOBAL_NAMES parameter using 'ALTER SYSTEM SET GLOBAL_NAMES=FALSE';
    • Using below example, create a db LINK ADB:
    CREATE DATABASE LINK DBL_TO_ADB 
       CONNECT TO SYEDJ IDENTIFIED BY welcome1 
       USING
    '(description=(retry_count=20)(retry_delay=3)
         (address=(protocol=tcps)(port=1522)(host=example1.oraclecloud.com))     
         (connect_data=(service_name=example2_high.adwc.oraclecloud.com))     
         (security=(my_wallet_directory=/u01/targetwallet)
            (ssl_server_dn_match=true)
            (ssl_server_cert_dn="CN=example2.oraclecloud.com,OU=Oracle
              BMCS US,O=Oracle Corporation,L=Redwood City,ST=California,C=US")))';

    At this point in time the DB link is ready.

    In nutshell, this blog post explained the do's and don'ts of creating DB link from/to ADB.





    Connecting to an Autonomous Database using JDBC Thin client

    I have discussed in my earlier posts about establishing SQL * Plus and using the SQLcl tool details. Today's blog post will take you through JDBC Thin client configuration details to be able to successfully connect with an autonomous database service. Will primarily focus on mandatory JDK, JDBC versions and client connect credentials requirement details. Let's get started.

    To begin with, any Java application that uses JDBC Thin driver requires an Oracle wallet or JavaKeyStore(JKS). Luckily, these files are readily available on OCI console for download. Refer my earlier blog post to learn about downloading client connect credentials files and JavakeyStore files.

    Prerequisites for JDBC Thin Driver

    Below is the list of perquisites for JDBC Thin client connection:

    • Download Connect Credentials as explained in the previous post
    • Validate JDK version for security, if you are on less than JDK8u162
    • Check JDBC driver version
    • JDBC driver 18.3 or higher simplifies security credentials wallet connection configuration.
    • Specify the wallet directory in the JDBC connect string.
    • Include the wallet directory location in the JDBC url if no TNS_ADMIN environmental variable is set.
    Connection string url

    Part of client connect credentials downloads, you will have tnsnames.ora files which contains tns service names and connection strings. Below is an example of high level connect string entry:

    dbname_high= (description=
          (address=(protocol=tcps)(port=1522)(host=adw.example.oraclecloud.com))(connect_data=(service_name=adw_jdbctest_high.oraclecloud.com))(security=(ssl_server_cert_dn="CN=adw.oraclecloud.com,OU=Oracle
          US,O=Oracle Corporation,L=Redwood City,ST=California,C=US")))
    JDBC 18.3 Thin Driver connection string example, for Linux platform:

    DB_URL="jdbc:oracle:thin:@dbname_high?TNS_ADMIN=/Users/test/wallet_dbname"

    Using JavaKeyStore (JKS)

    Below example shows the JDBC Thin driver connectivity using JKS to an ADB:

    DB_URL="jdbc:oracle:thin:@dbname_high?TNS_ADMIN=/Users/test/wallet_dbname"
    Subsequently, add the JKS related connection properties to ojdbc.properties file. To use SSL connectivity over Oracle Wallet, below properties should be set to ojdbc.properties files:


    # Properties for using Java KeyStore (JKS)
    oracle.net.ssl_server_dn_match=true
    javax.net.ssl.trustStore==${TNS_ADMIN}/truststore.jks
    javax.net.ssl.trustStorePassword=password
    javax.net.ssl.keyStore==${TNS_ADMIN}/keystore.jks
    javax.net.ssl.keyStorePassword=password

    Once the properties are set, compile and run using the below example:

    java –classpath ./lib/ojdbc8.jar:./lib/ucp.jar UCPSample
    To go through your firewall with HTTP proxy requirement to connect to an internet, use JDBC Thin client 18.1 or higher. So, add the following to your tnsnames.ora file to go through HTTP proxy:

    ADWC1_high =
           (description=
                 (address=
                       (https_proxy=proxyhostname)(https_proxy_port=80)(protocol=tcps)(port=1522)(host=adw.example.oraclecloud.com)
                 )
                 (connect_data=(service_name=adwc1_high.adw.oraclecloud.com)
                 )
                 (security=(ssl_server_cert_dn="adw.example.oraclecloud.com,OU=Oracle BMCS US,O=Oracle Corporation,L=Redwood City,ST=California,C=US")
                 )
           )

    In nutshell, we have walk through over JDBC thin driver client connection to an ADB requirements and prerequisites.

    4.03.2020

    Autonomous Database (ADB) connectivity with Oracle SQLcl

    This blog post takes you through Oracle SQL Developer command line interface connectivity to an autonomous database (ADB). SQLcl is a free command line interface for Oracle database. It will allow interactive or batch execution of SQL and PL/SQL commands with in-line editing, statement completion, and command recall for a feature-rich experience capabilities. Download the tool from the below url:



    With SQLcl 4.2 or higher you can also connect  to ADB database, either using Oracle Call Interface or JDBC thin connection. Whatever connection mode, OCI or JDBC, it has to be setup before you can use the SQLcl.

    Use the below example to connect to a ADB using SQLcl oci connection:

    sql -oci
    
    SQLcl: Release 18.4 Production on Wed Apr 03 15:28:40 2019
    
    Copyright (c) 1982, 2019, Oracle.  All rights reserved.
    
    Username? (''?) sales_dwh@adwc1_low
    Password? (**********?) **************
    Last Successful login time: Wed Apr 03 2019 15:29:19 -07:00
    
    Connected to:
    Oracle Database 18c Enterprise Edition Release 18.0.0.0.0 - Production
    Version 18.4.0.0.0
    
    SQL>
    Use the below example to use SQLcl with JDBC thin connection:

    • Start with sql /nolog 
    • Configure the session to use existing wallet connection: 
    • SQL> set cloudconfig /home/adwc/wallet_ADWC1.zip
      Wallet Password:  **********
    • Connect to ADB database , as shown below:
    • SQL> connect admin@adwc1_medium

    In my next post, will discuss JDBC connectivity details.

    SQL * Plus connectivity to Autonomous Database (ADB) procedure

    Autonomous Database (ADB) learning keeping myself pretty occupied these days. All credit should go to Oracle for their free offering of Oracle cloud technology university courses and a chance to certify. So planned to write a few important topics along the way while learning about ADB. This blog post will primarly focus on autonomous database (ADB) service available connectivity options, downloading database credentials and a procedure to connect to an ADB database service using the SQL * Plus tool.

    An Autonomous Database (ADB) service can be connected through SQL * Net, JDBC thick/thin, ODBC processes etc. ADB connection uses certificate authentication and SSL, where  certificate authentication uses encrypted key stored in a wallet files on both client and server. All client connections to the database must use SSL for encryption and under any circumstances it doesn't accept any unsecured connections. 

    To begin with, every ADB database connection must be made over a public internet and application should use a secure connection (SSL).  Before establishing any sort of client connectivity to the ADB service, service client connect credentials must be downloaded from the OCI page for the respective database (instance) or region. Below procedure summarizes how to download database connect (wallet) credentials and use of SQL * Plus to connect to  ADB.


    Downloading database connection credentials

    As discussed earlier, ADB only accepts secure connections to an Autonomous Database. Therefore, Oracle client credential (wallet) files which contains your connection credentials should be downloaded first. You need to download this files only once. The wallet can be downloaded either from Administration link from the ADB service console page, or from the DB Connection button on instance page of OCI page or through API calls. Use the following method to download the connection credentials from OCI console:

    • Navigate to ADB details page  and click on DB Connection button
    • Based on the connectivity requirement, choose either Instance or Regional wallet type and click on the Download button 
    • The credentials file is protected with a password. So, you will have to input a password before downloading the file. The password must be 8 characters long, with at least 1 letter and either 1 numeric character or 1 special character
    • Once the password is entered and confirmed, click the download button
    • Default name for the file is Wallet_databaename.zip, and the zip files contains following files:
      • tnsnames.ora & sqlnet.ora
      • cwallet.sso & ewallet.p12
      • keystore.jks & tuststore.jks
      • ojdbc.properties & 
    • Unzip the file to a secure location
    Predefined DB Service Names

    The tnsnames.ora files contains the following predefined 5 database services names for Autonomous Transaction Processing (ATP) and 3 (high, medium, low) for Autonomous Data warehouse (ADW) databases:
    • high A high priority app connection for reporting and batch operations which runs in parallel with little concurrency capacity, subject to queuing.
    • medium A typical connection service for reporting and batch operations with limited (4) parallel degree.
    • low A low priority connection service for reporting and batch operations with no parallelism options.
    • tp (for ATP only), typical application connection service for transaction processing operations with no parallelism.
    • tpurgent (ATP only) the highest priority application connection for time critical transaction processing operations. Can specify manual parallelism. 
    By default, sessions on ADB that are idle for over 5 minutes will be automatically disconnected from the database to make room for more active sessions.

    SQL * Plus connection to ADB

    For any sort of ADB connection, database credentials must be dowloaded by an individual user first. In this segment, we will discuss how to use an SQL * Plus tool to connect to ADB to execute regular SQL commands. Unlike other Oracle cloud options, ADB doesn't provide compute node access. So, you must have SQL * Plus installed on your computer. So, you have to install Oracle client software 11.2.0.4 or higher version on your system. Once the client software is installed, download connection credentials using the procedure described above.

    Unzip/uncompress the zip file downloaded into a secure folder on your computer.

    Edit the sqlnet.ora file and replace ?/network/admin text with name of the folder containing the credentials, as shown in the below example on a unix/linux client:

    WALLET_LOCATION = (SOURCE = (METHOD = file) (METHOD_DATA = (DIRECTORY="/home/adw_credentials")))
    SSL_SERVER_DN_MATCH=yes

     It is highly recommended to set TNS_ADMIN environmental variable and set the location of the credentials files.

    As previously stated, the connection must be through an internet. So the companies that doesn't provide direct internet connection, they will allow the access through HTTP proxy settings. In this case, you must update the sqlnet.ora and tnsnames.ora files as per below example if you have a firewall connection and requires HTTP proxy to connect internet (this scenario only support clients with 12.2.0.1 or higher version):

    • sqlnet.ora should have this entry: SQLNET.USE_HTTPS_PROXY=on
    • And the tnsnames.ora entry should contains the following:
    ADWC1_high =
           (description=
                 (address=
                       (https_proxy=proxyhostname)(https_proxy_port=80)(protocol=tcps)(port=1522)(host=adwc.example.oraclecloud.com)
                 )
                 (connect_data=(service_name=adwc1_high.adwc.oraclecloud.com)
                 )
                 (security=(ssl_server_cert_dn="adwc.example.oraclecloud.com,OU=Oracle BMCS US,O=Oracle Corporation,L=Redwood City,ST=California,C=US")
                 )
           )
    Once you have the client software and connect credentials ready, now you can use the sql * plus tool to connect to a ADB. Refer below syntax using low credentials service name here:

    sqlplus sales_dwh@adwc1_low  
    
    SQL*Plus: Release 18.0.0.0.0 - Production on Wed Apr 3 15:09:53 2019
    Version 18.5.0.0.0
    
    Copyright (c) 1982, 2018, Oracle.  All rights reserved.
    
    Enter password:
    Last Successful login time: Wed Apr 03 2019 14:50:39 -07:00
    
    Connected to:
    Oracle Database 18c Enterprise Edition Release 18.0.0.0.0 - Production
    Version 18.4.0.0.0
    
    SQL>
     You can also see the last successful login time.

    In nutshell, this post explained how to download ADB database connect credentials from various methods, how to connect to ADB using SQL * Plus tool. Will be discussing about JDBC thin client connectivity procedure in the next post.


    References;
    https://docs.oracle.com/en/cloud/paas/autonomous-data-warehouse-cloud/user/connect-sqlplus.html#GUID-A3005A6E-9ECF-40CB-8EFC-D1CFF664EC5A