How to find users who didn't activate MFA on OCI

Today this came up, I needed to find all the users in a certain Group in my identity domain. Of course console provides this information, if I look at details of all users one by one! But I don't want to do this, it is time consuming, error prone and doesn't scale. Imagine having hundred users!

1 OCI CLI should also provide this information. But not probably with one API call, so let's look at the following script:

First I am trying to queryonly the OCID of the Group with name"Administrators"

oci iam group list --compartment-id $TENANCY_OCID --name Administrators \
 --query 'data[*]."id"' --raw-output

Then after couple of string operations to eliminate paranthesis and double quotes. I am passing the output of the first command which is the OCID of Administrators Group to a second command. This time to find the users with a specific group assignment, and only interested in id and name columns. And a table like output would be better compared to default JSON.

oci iam group list-users --compartment-id $TENANCY_OCID --group-id $GROUP_OCID \
 --query 'data[*].{OCID:id,Name:name}' --output table

As you can see, it is quite handy and reliable. You can check Christian Gohmann's post for neat and tidy samples, link in the references.

2 CLI is good yet not the easiest way though. Especially when combining multiple commands, it requires a certain level of expertise on scripting. So is there an easier solution? Sure there is: Steampipe Since my colleague Jean-Pierre showed this, I am loving it. It has an OCI Plugin , and here is the GitHub Page of the plugin code. Once you install and run it, you will see it makes your day easy as joining couple of tables with an SQL is.

It supports PostgreSQL syntax, and using plugins you can use it for AWS, Azure or OCI. Or you can write your own. Here is another query that I needed. List of users who didn't activated MFA with last login time:

Steampipe Plugin documentation has hundreds of queries provided for different use cases.

Note: From this point on I am sharing my notes on how to do things

Install OCI CLI: On my WSL2 Ubuntu system, I needed to re-install cli. Quickstart is quite handy, script is doing everything for you. Once you install it, you need to configure it by following the prompts:

oci setup config

Extending Steampipe OCI Plugin: I realized the Last Login Date data was not available on plugin table but available on oci cli. So I decided to extend the plugin code. For this purpose:

1.I cloned the git repository as described in GitHub Page .

git clone https://github.com/turbot/steampipe-plugin-oci.git
cd steampipe-plugin-oci

2. I added last_successful_login_time into table definition file table_oci_identity_user.go
3. Installed Go on my Ubuntu host, and added to my path

rm -rf /usr/local/go && tar -C /usr/local -xzf go1.21.1.linux-amd64.tar.gz
export PATH=$PATH:/usr/local/go/bin

4. Then build the plugin from source and configure it as described on plugin GitHub page

make
cp config/* ~/.steampipe/config
vi ~/.steampipe/config/oci.spc

References:
1. OCI CLI Search and Filtering: https://christian-gohmann.de/2021/03/10/forma...
2. Steampipe: https://hub.steampipe.io/plugins/turbot/oci
3. Steampipe OCI Plugin: https://github.com/turbot/steampipe-plugin-oci
4. OCI CLI Installation: https://docs.oracle.com/en-us/iaas/Content/API/SDKDocs/cliin...
5. Install Go: https://go.dev/doc/install
6. Steampipe OCI Plugin Sample Queries: https://hub.steampipe.io/mods/turbot/oc...

Run Autonomous Database, ORDS and APEX on your laptop with single command!

Last week was full of excitement, you know it is that time of the year: Cloud World You can watch the recaps! Lots of announcements, new partnerships, product launches, demos, tons of interesting sessions and chance to connect with gurus, product managers and community! I didn't have the chance to be there yet, maybe next year...

One of the announcements was a container image for Autonomous database made available! It has built-in tools like Database Actions (SQL Developer Web, Performance Hub, etc.), ORDS and APEX, and Mongo API is enabled. Just the right things for developing locally without loosing anytime. Here is the offical documentation and the GitHub Page where you can find all the details.

So here is what I did to have my container running on my Windows laptop within WSL2 Ubuntu.

1We start with podman installation (you can also use docker)

2When the container runs the following ports will be exposed:

Port	Description
1521	TLS
1522	mTLS
8443	HTTPS port for ORDS / APEX and Database Actions
27017	Mongo API ( MY_ATP )

I recommend pulling the image first. Size is around 10GB and it can take a while. You can run the container with the following command.

3 Now we need to change ADMIN user password. There is a script provided for this purpose and we need to execute it by connecting to container.

4 We are ready to explore the tools already provided. Point your browser to https://localhost:8443/ords/my_atp/ and a landing page will welcome us.

5 APEX and Database Actions are also made available, no installation, no configuration, start building immediately.

6 How about connecting to database? Easy, for mTLS it requires a wallet. You can copy the wallet to any location on your local filesystem, export TNS_ADMIN then connect.

Note: You can safely skip this first part, unless you want to update your WSL2 Ubuntu. I was using a manual built experimental kernel because of a really weird debugging requirement I had in the past and I didn't need it anymore. So I needed to replace it but never had the chance or motivation, but this time it was inevitable. So writing this section as a reference for my future self.

References:
1. ADB Free GitHub Page: https://github.com/oracle/adb-free
2. Podman Documentation: https://podman.io/docs/installation
3. Instal Docker on WSL2: https://nickjanetakis.com/blog/install-docker-in-wsl-2-without-do...
4. Autostart Docker Daemon: https://blog.nillsf.com/index.php/2020/06/29/how-to-autom...
5. Upgrade Ubuntu: https://askubuntu.com/questions/1428423/upgrade-ubuntu-in-wsl...
6. Reboot Ubuntu: https://superuser.com/questions/1126721/rebooting-ubuntu-o...

How to discover tenancy details and self OCID with Autonomous database

Know thyself inscribed on Temple of Apollo, it is the door to true wisdom. While scripting or coding for automating a process, I often need the database to be self aware and discover information like tenancy, OCID and region etc. So here is how you can obtain details on Autonomous database

cloud_identity column contains a json formatted text containing OCIDs for tenancy, autonomous database, compartment etc. And if you are scripting/coding you will need to extract those json attributes, you can use JSON_TABLE function. This will give you nice and clean values that you can directly use.

How to connect Autonomous Database using java JDBC with or without a wallet

Sometimes it is good to go back to the basics. Connection between Autonomous database and the client is always encrypted with TLS. Depending on the network access type, either it has to be mutual (mTLS) or TLS only. Why this is important? Well, I can think of couple good reasons. If I allow "secure access from everywhere" to my database then I would want to ensure only the clients that I shared necessary information (in this case a wallet) should be able to access it. Other logical options are allowing access only from the network addresses that I know, like my own IP address or from private network that I trust.

So I will be writing a small piece of java code to test it. It is all about setting my environment and constructing JDBC connection string. So at the minimum I add jdbc driver to my pom file. You might want to use UCP for your applications, it will be the same. I will be following official documentation

1 In order to connect with a Wallet, I download the wallet from console, unzip it then use the folder as my TNS_ADMIN. Connection string looks like

jdbc:oracle:thin:@{CONNECT_DENTIFIER}?{TNS_ADMIN}

As you will see there are different ways to construct it: Wallet folder can be set as a system property or passed as a query string. Here is a small sample code

2 If I can restrict access to my autonomous database with an access controll list (ACL) or connect with a private endpoint in a private subnet then I can uncheck Require mutual TLS (mTLS) authentication checkbox then I can connect to this database without a wallet. Connection is still encrypted with TLS.

Tracking Cloud Usage and Cost with Autonomous Database Using SQL

This was long time in my waiting list, "How do I track usage/cost of my cloud resources?" A very common question that I hear a lot. Let me walk through the options I know:

1 I can use cost analysis tool. It is very handy for visualizations, I can add filters, change grouping dimensions. I can have daily and monthly reports including forecast for the interested period. I can also save my reports, download report output in different formats. I can even schedule reports and it will be created under a bucket. Down sides are: it is built for visual consumption, extracting data is manual, I can't go beyond one year.

2 I can use cost and usage reports. They are csv files generated every six hours in an Oracle owned bucket but can be accessed with some cross tenancy configuration. These files are retained for one year. See official documentation for details. I need to find other tools to import and analyze data in csv files. Here is a really good example for this purpose.

3 I can also use the REST API to get a Usage Summary. Similar is also available in cli. Depending on the requirement this might be a good fit, but most of the time it is too much for most of the customers.

4 I can configure an Autonomous Database and use it to Track Oracle Cloud Infrastructure Resources, Cost and Usage Reports with Autonomous Database Views . I am going to focus on this because with minimal effort it can deliver great value. I can query data with simple sql, integrate it into any reporting or monitoring tool and I can go beyond one year limit by storing csv report data inside my database. Here I will list steps and some sample scripts:

a There are some prerequsite steps that need to be completed. You can check official documentation or my other blog post where I discuss resource principals to access OCI resources.

Note:The resource principal token is cached for two hours. Therefore, if you change the policy or the dynamic group, you have to wait for two hours to see the effect of your changes. This note is from documentation .

b Once the policies and resource principals are in place, I can use resourse views. There are multiple views but I am interested in two of them: OCI_COST_DATA and OCI_USAGE_DATA. I didn't check the source but I am guessing PIPELINED functions and/or external tables are involved. I could just use the views but there are two problems I need to solve: running query on the view is slow, and underlying data is changing, data older than one year will vanish. For this reason I am going to create a materialized view based on an existing table. This temporary table will be refreshed every 6 hours and in an another table I will be accumulating all the data. I am creating primary keys to detect the difference for each run.

c I am creating a procedure which will refresh the temp table with latest data, merge new rows into actual table, then schedule the next run.

d I just need to execute the procedure once, then the ball will be rolling on its own.

e DBMS_SCHEDULER job details can be tracked with the following views.

f Finally I will have all the data accumulated in my table, even long after the csv usage/cost files are deleted. Here are some SQL queries to start with.

Different Ways to Access Cloud Resources from Autonomous Database

When working with DBMS_CLOUD package or cloud REST APIs , I need database instance to be authenticated and authorized. Mainly there are two ways of doing this.

1I can use my own credentials or any IAM users credentials. For this purpose I need to use DBMS_CLOUD.create_credential procedure that comes in three different signatures.

aI can create an Auth token from console or using cli

then using this token and my user I can create a credential. Just to avoid confusion with below script, I use my email address as username in my tenancy.

bAnother way is to introduce my API signing RSA keys to OCI, then use it to create a credential. For generating my own key pair I can use openssl as described here in official documentation . I can also use the console which can generate the keys for me and I can download it. Using console I can upload my existing keys too.

After the API key added to OCI, console will display a configuration that can be used with SDK, CLI or REST calls.

CLI doesn't offer a command for adding API keys but I can always use REST API with http raw request, again response will display required information to use API key with SDK and CLI

Note:Use \n as new line feed for formatting your encoded public/private key

Now I can use a different version of create_credential procedure

Note:Both credentials (Auth Token and API Key) are directly linked to my OCI IAM user.

2I can also use Resource Principals to authorize my ATP instance. Previous method is tied to an IAM user (notice both Auth Token and API Key are created under user), resource principal uses Dynamic Groups to identify the instance and IAM no user is required.

aFirst I need a Dynamic Group to identify my instances. I generally use tagging, but sometimes allowing all autonomous instances is also fine.

bThen with a policy I grant priviliges to the members of that dynamic group

Note:The resource principal token is cached for two hours. Therefore, if you change the policy or the dynamic group, you have to wait for two hours to see the effect of your changes. This note is from documentation .

Here is the complete list of cli commands with some outputs for the same purpose:

cAnd I connect to the database and enable Resource Principal to Access Oracle Cloud Infrastructure Resources .

Testing

1I can see that my credential is visible and enabled in all_credentials. For testing I am just listing objects under an object storage bucket

2I can list objects under a bucket using any of the credentials.

Here is some SQL for testing

Putting it altogether: How to deploy scalable and secure APEX on OCI

Oracle APEX is very popular, and it is one of the most common usecases that I see with my customers. Oracle Architecture Center offers a reference architecture for this purpose: Deploy a secure production-ready Oracle Autonomous Database and Oracle APEX application . If you are comfortable with terraform or willing to learn it, I would definitely recommend using it. Even if it doesn't fit your requirements entirely it is a good starting point.

Having said that, I wanted to crack it open and see what's in it (I know typical boy's fun...), and in the end I come up with a series of blog posts while building the reference architecture piece by piece. Good for understanding what is under the hood and excellent for showing the value of terraform after doing all the work manually.

So here I start with the final architecture. I will explain the components and provide the links while doing so to help you build your own.

Quick links to the posts in the series

Part 1: Accessing Autonomous Database over Private Endpoint using SQL Developer
Part 2: Installing Customer Managed ORDS on Compute Instance for Autonomous Database
Part 3: Serving APEX in Private Subnet behind Public Load Balancer
Part 4: Securing APEX Admin Resources with Load Balancer
Part 5: Autoscaling ORDS servers in private subnet behind public load balancer

1Backbone of everything is Oracle database for APEX applications. I have an Autonomous database instance with Transaction processing workload type with autoscaling enabled. It is deployed with a Private Endpoint in a private subnet. You can check official documentation for creating one. For accessing your ATP instance you can see Part 1: Accessing Autonomous Database over Private Endpoint using SQL Developer

2Although ATP comes with Oracle managed ORDS, I want to install my own ORDS server on compute vm. In Part 2: Installing Customer Managed ORDS on Compute Instance for Autonomous Database I install and configure java and ORDS, also do the required networking configuration.

3For improving security posture, both database endpoint and ORDS instance is placed in a private subnet. For exposing APEX application I follow the steps in Part 3: Serving APEX in Private Subnet behind Public Load Balancer . This part is all about load balancer configuration, backend health check, SSL termination and troubleshooting connection issues. It can be helpful for any kind of load balancer / application configuration and problem solving.

4In a real life deployment, I need to find a way to access admin resources yet be able to protect them from public internet access. For this purpose, I am securing some URLs with load balancer redirect rules as load balancer sits in between as a reverse proxy. I can still access those admin resources through private subnet using Fastconnect, VPN or bastion service. These topics covered in Part 4: Securing APEX Admin Resources with Load Balancer

5Autonomous database will scale up to 3x according to CPU requirements, that is easy configuration. For the middleware part, I use metrics based auto scaling for adding ORDS instances when existing instances in the pool have 80% or more CPU utilization. I cover scaling configuration along with testing in Part 5: Autoscaling ORDS servers in private subnet behind public load balancer

I also recommend checking my colleague John Lathouwers's GitHub , he has some nice scripts.