Setup RDF Graph Server with Autonomous Database

Quick start:

• Prerequisites
• Create Autonomous Database
• Create Network for RDF Graph Server
• Create RDF Graph Server
• Download Wallet
• Modify Wallet
• Upload Wallet
• Create RDF Network
• Run SPARQL Query

Appendix:

• Change RDF Server Password
• Load Triples with Graph Name

Quick Start

Prerequisites

• Oracle Cloud account
• 1 SSH key pair (private / public key)
• 3 passwords
  • <password_1>: DB login (ADMIN user)
  • <password_2>: DB Connection Wallet
  • <password_3>: RDF Server login (weblogic user)

Create Autonomous Database

Oracle Cloud console > Oracle Database > Autonomous Database > Create Autonomous Database

• Configure the database
  • Database name: <db_name> (e.g. ADB1)
  • Database version: 19c or 21c
  • Password: <password_1>

Create Network for RDF Graph Server

Oracle Cloud console > Networking > Virtual Cloud Networks > Start VCN Wizard > VCN with Internet Connectivity

• Configuration
  • VCN NAME: (e.g. VCN1)
  • The rest of itmes: Do not need to change

Public Subnet vcn01

• Add Ingress Rules
  • Source CIDR: 0.0.0.0/0
  • Destination port range: 7002,8001
  • Description: For RDF Graph Server

Create RDF Graph Server

Oracle Cloud console > Marketplace > All Applications

• Image
  • Name: Oracle RDF Graph Server and Query UI
  • Version: 21.3.0.1
• Configure Variables
  • Server available domain: Any domain
  • Server shape: Any shape
  • SSH public key: Paste the content of the public key (e.g. ssh-rsa AAAAB3NzaC...)
  • Existing virtual cloud network: The VCN created above
  • Existing subnet: The public subnet of the VCN

Please check the IP address of this compute VM via the Web console.

Download Wallet

Oracle Cloud console > Oracle Database > Autonomous Database

Select the ADB created above > DB Connection

• Wallet Type: Instance Wallet

Download

• Password: <password_2>

Modify Wallet

Upload the wallet to the RDF Server and run a script to add the user name and password to the wallet.

$ scp -i <key> ~/Downloads/Wallet_<db_name>.zip opc@<ip_address>:
$ ssh -i <key> opc@<ip_address>
$ mkdir wallet
$ cd wallet
$ unzip ../Wallet_<db_name>.zip
$ export JAVA_HOME=/usr/local/java/jdk1.8.0_221
$ /u01/app/oracle/middleware/wls12214/oracle_common/bin/mkstore
  -wrl /home/opc/wallet -createCredential <db_name>_high ADMIN <password_1>
Enter wallet password: <password_2>

Example:

$ scp -i .ssh/id_rsa ~/Downloads/Wallet_ADB1.zip opc@192.0.2.1:
$ ssh -i .ssh/id_rsa opc@192.0.2.1
$ mkdir wallet
$ cd wallet
$ unzip ../Wallet_ADB1.zip
$ export JAVA_HOME=/usr/local/java/jdk1.8.0_221
$ /u01/app/oracle/middleware/wls12214/oracle_common/bin/mkstore \
  -wrl /home/opc/wallet -createCredential adb1_high ADMIN <password_1>
Enter wallet password: <password_2>

Download the modified wallet.

$ zip ../wallet_with_cred.zip *
$ exit
$ scp -i <key> opc@<ip_address>:~/wallet_with_cred.zip ~/Downloads/

Example:

$ zip ../wallet_with_cred.zip *
$ exit
$ scp -i .ssh/id_rsa opc@192.0.2.1:~/wallet_with_cred.zip ~/Downloads/

Upload Wallet

Sign in to the RDF Graph Server and Query UI

• URL: https://<ip_address>:8001/orardf
• User: weblogic
• Password: welcome1 (This password should be changed in the latter step)

Data sources tab > Create > Wallet

• Zip file: Select wallet_with_cred.zip
• Name: Any name of this data source (e.g. ADB1)
• Wallet Service: The service modified above <db_name>_high (e.g. adb1_high)

Create RDF Network

Data tab

• Data source: The data source created above (e.g. ADB1)

Data tab > RDF Network > Create network icon

• Network owner: ADMIN
• Network name: Any network name (e.g. NETWORK1)
• Tablespace: DATA

Data tab

• RDF network: The RDF network created above (e.g. ADMIN.NETWORK1)

Data tab > Import > Upload data icon

• Upload: Select sample.nt
• Staging table: Input any new table name (e.g. SAMPLE_TABLE)
• Overwrite: Off (unless the table name is used already)

Data tab > Import > Bulk load data icon

• RDF Data
  • Model: Input any model name (e.g. SAMPLE_MODEL)
  • Staging table owner: ADMIN
  • Staging table: Select the table created above (e.g. SAMPLE_TABLE)
• Options
  • All of the items: Do not need to change
• Event Trace
  • All of the items: Do not need to change

Run SPARQL Query

Data tab > Directory Tree > RDF Objects > Regular models

Right click the model (e.g. SAMPLE_MODEL) > Open

Click Execute button to run the SPARQL query in the text box.

From a new browser tab, access the URL below to send a SPARQL query as a GET request.

https://<ip_address>:8001/orardf/api/v1/datasets/query?datasource=ADB1&datasetDef={"metadata":[{"networkOwner":"ADMIN","networkName":"NETWORK1","models":["SAMPLE_MODEL"]}]}&query=select ?s ?p ?o where { ?s ?p ?o} limit 10

Appendix

Change RDF Server Password

Access WebLogic Server console using Firefox.

• https://<ip_address>:7002/console

Login with the default password.

• Username: weblogic
• Password: welcome1

Go to Security Realms > Users and Groups, and select weblogic.

Change password from the Passwords tab.

Login to RDF Graph Server console with the new password.

Load Triples with Graph Name

From RDF Server, upload data to a staging table as explained at Create RDF Network.

• Upload: Select sample.nt
• Staging table: Input any new table name (e.g. SAMPLE_TABLE)
• Overwrite: Off (unless the table name is used already)

Add a new column for a graph name to the staging table as follows.

Data tab > Import > Bulk load data icon

Oracle Cloud console > Oracle Database > Autonomous Database

Select the ADB created above > Tools > Open Database Actions

create view SAMPLE_VIEW as
select
  RDF$STC_SUB
, RDF$STC_PRED
, RDF$STC_OBJ
, '<http://example.org/graph1>' as RDF$STC_GRAPH
from SAMPLE_TABLE

Go back to RDF Server and import from the staging table (= a new view SAMPLE_VIEW) to a model.

Data tab > Import > Bulk load data icon

• RDF Data
  • Model: Input any model name (e.g. SAMPLE_MODEL)
  • Staging table owner: ADMIN
  • Staging table: Select the table created above (e.g. SAMPLE_VIEW)
• Options
  • All of the items: Do not need to change
• Event Trace
  • All of the items: Do not need to change

Now this SPARQL query should return the graph name.

SELECT *
WHERE { GRAPH ?g { ?s ?p ?o }} 
LIMIT 500