Category Archives: Database

Migrating Your Home Assistant Database from SQLite to PostgreSQL

Migrating your Home Assistant database from SQLite to PostgreSQL can significantly enhance performance, especially as your data grows. This guide will walk you through creating a database dump, converting data types, setting up your PostgreSQL database, and configuring Home Assistant to use the new database.

Creating the Database Dump from SQLite

To start, you need to create a dump of your existing SQLite database:

sqlite3 home-assistant_v2.db .dump > ha_dump.sql

This command creates a plain text file containing the SQL commands needed to reconstruct the database.

Converting Data Types

During the migration from SQLite to PostgreSQL, certain data types need to be converted:

sed -i 's/DATETIME/TIMESTAMP/g' ha_dump.sql
  • BLOB to BYTEA: Convert BLOB fields to BYTEA for binary data.
sed -i 's/BLOB/BYTEA/g' ha_dump.sql

Preparing PostgreSQL Database

Creating Database and User

Start by setting up your PostgreSQL database and user:

CREATE DATABASE homeassistant;

Create the objects and load the data

Use the command line utility psql to create the database objects and load the data into the newly created PostgreSQL database:

psql -h [your_db_host]-U ha -d homeassitant -f ha_dump.sql -W > load.log 2>&1


  • Replace [your_db_host] with the actual hostname or IP address where your PostgreSQL database is hosted.
  • User and Database: Ensure that ha is the correct username and homeassistant is the correct database name you created for Home Assistant.

After running the command, the SQL dump file (ha_dump.sql) will be executed against your PostgreSQL database. The output and any errors encountered during the process will be redirected to load.log. This log file is essential for tracking the progress and identifying any issues that need resolution.

Check for Errors:

  • Review the load.log file for possible errors. This file contains all output from the psql command, including any SQL errors or warnings that were generated during the import process.
  • Iterate as Necessary: If errors are found, you may need to fix issues in ha_dump.sql and rerun the command. This step might need to be repeated several times. Modifications could involve correcting data types, adding missing sequences for auto-incrementing fields, or adjusting SQL syntax to be compatible with PostgreSQL.

This process can be time-consuming, but it is crucial for ensuring that your database is correctly set up with all the necessary data and schema configurations. As noted, your mileage may vary depending on the specifics of your data and the initial state of the ha_dump.sql file.

Adjusting the Schema

For tables needing auto-increment functionality (which is common in primary key columns), set up sequences:

CREATE SEQUENCE states_state_id_seq;
ALTER TABLE states ALTER COLUMN state_id SET DEFAULT nextval('states_state_id_seq');
SELECT setval('states_state_id_seq', (SELECT MAX(state_id) FROM states) + 1);

Repeat this pattern for other necessary tables and columns, such as events(event_id), state_attributes(attributes_id), and so on.

Configuring Home Assistant

Install SQLAlchemy and other dependencies

Home Assistant uses SQLAlchemy as the SQL toolkit and Object-Relational Mapping(ORM)system for Python. Install it with pip:

pip3 install SQLAlchemy<br>pip3 install psycopg2-binary

Modify the configuration

Modify the configuration.yaml file to point to the new PostgreSQL database:

  db_url: postgresql://ha:yourpassword@localhost/homeassistant

This setup directs Home Assistant to use the newly configured PostgreSQL database.

Benefits of Migrating to PostgreSQL

Moving from SQLite to PostgreSQL offers several benefits:

  • Scalability: PostgreSQL handles larger databases and more concurrent connections.
  • Performance: Improved query performance and optimization options.
  • Reliability: Robust transaction support and recovery features.
  • Flexibility: Richer set of data types and full-text searching capabilities.


Migrating your Home Assistant database to PostgreSQL not only enhances performance but also provides a more robust and scalable backend, suitable for growing smart home environments. This migration ensures that your Home Assistant setup can handle increased data loads efficiently and reliably.

Famous last words

One of my primary concerns during this migration was the potential loss of historical data, particularly how it might affect critical metrics like energy usage. The statistics table, which was the last to have the auto-incremental column added, is pivotal as it houses the energy usage stats.

As the image below shows, there appears to be a gap of approximately four hours in the data on energy usage stats. However, it seems that Home Assistant has effectively compensated for this missing data. The system appears to have aggregated the missing energy usage from those four hours into the data represented in the 5 PM bar on the chart.

This outcome is quite reassuring and confirms that the system’s integrity remains intact despite the migration hiccups. I’m relieved to see that after all the adjustments and troubleshooting, everything is functioning as expected.

This experience underscores the importance of careful planning and execution in database migrations, especially when dealing with essential home automation systems like Home Assistant. The transition may require significant effort and attention to detail, but the end result can be gratifying, ensuring continuity and robustness in data handling.

Building a DIY AI Chatbot: Control Your Conversations


A self-built AI chatbot is crafted entirely by an individual or team from scratch, without relying on pre-existing templates or platforms. This approach gives developers complete autonomy over the coding, features, and functionalities of the chatbot.

Creating a self-built AI chatbot demands a blend of programming expertise, a deep understanding of artificial intelligence, and inventive thinking. Developers can use a variety of programming languages, including Python, Java, or JavaScript, based on their preferences and the chatbot’s intended application.

One of the standout advantages of a self-built AI chatbot is its high level of customization. Developers can fine-tune the chatbot’s responses and functionalities to meet specific needs and objectives. Moreover, they can continually refine and enhance the chatbot on their own timetable, independent of external updates or support.

Getting started

Building a chatbot from scratch might seem daunting, but it’s quite feasible with the right tools. I used the OpenAI API and the Python openai library (version 1.23.2 as of this writing). While GPT-4 typically suggests using openai==0.28, the transition to versions above 1.0 signifies substantial changes and necessitates thoughtful consideration. However, this doesn’t mean that ChatGPT cannot assist in coding—it can, though it requires precise instructions.

Technical setup

For my project, the technical foundation included:

  • Python 3.9.x or higher: I chose Flask as the application server.
  • Access to the OpenAI API: Essential for integrating the AI logic into the chatbot

This setup is sufficient to establish a testing environment for the AI logic, connecting the Python code to the OpenAI API.

Advanced configuration

After thorough testing, I moved on to production. I continued using Flask for its simplicity, but also added Gunicorn as a frontend server. The application runs either as a standalone version or embedded within a WordPress blog.

I explored different operational models, including storing interactions in a database and the Bring Your Own Data (BYOD) model, although the latter’s impact on performance is still unclear. Initially, I deployed the gpt-3.5-turbo-instruct model for its speed and contextual retention. However, for superior output quality, I ultimately chose GPT-4 despite its slower response time.

The AI Bot Herself

The embedded ChatBot is utilizing gpt-3.5-turbo-instruct whereas the one on below links is utilizing gpt-4 model. The later needs a bit time to think, but she will get there… You can compare the results.


A self-built AI chatbot can serve myriad purposes—customer support, entertainment, educational assistance, or personal aid, and can be integrated across websites, messaging platforms, or mobile apps.

For me, the project was primarily an exploration of AI technologies and the OpenAI API. It was also an invaluable learning experience in Python, application servers, and container technologies.

Building a self-built AI chatbot is undoubtedly a complex, resource-intensive endeavor that necessitates ongoing updates and maintenance. Yet, the potential for continuous learning and improvement through natural language processing and machine learning algorithms makes it increasingly efficient and precise over time.

From a Friday morning start to a productive Monday evening, my journey with this project underscores the potential and versatility of AI technologies, making a self-built AI chatbot a potent, customizable tool for any tech-driven initiative.


Streamlining Your Database Migration: A Guide to Leveraging OpenAI API for Seamless Assessments

Database migration is a complex process that demands careful assessment to ensure data integrity, application performance, and overall system reliability. The OpenAI API, with its advanced natural language processing capabilities, offers a way to simplify this process by automating assessments and summarizing key points. This guide will walk you through using the AWS Schema Conversion Tool (AWS SCT) for initial assessments, integrating the OpenAI API with Python to generate assessment summaries, and understanding the requirements for connecting with Azure OpenAI API, as well as its differences from ChatGPT OpenAI.

Kickstarting Your Migration: Utilizing AWS SCT for Comprehensive Database Assessment

The Amazon Web Services Schema Conversion Tool (AWS SCT) simplifies database migration from one platform to another. It assesses your existing database schema and generates a detailed report on potential migration issues. Supporting a wide range of source and target databases, AWS SCT is versatile for many migration scenarios.

AWS SCT examines your database schema, identifies non-convertible elements, and produces a comprehensive report. This report, containing potential action items, is crucial for planning your migration, offering an overview of the complexity, potential challenges, and the effort required.

The report, in PDF format, provides a detailed view of your database schema, potential issues, and recommendations. While invaluable for database administrators and engineers, the report’s extensive and complex nature makes OpenAI API a perfect tool for simplification and summarization.

Transforming PDFs into Comprehensive Assessment Summaries

With the AWS SCT report in hand, the next step is to utilize OpenAI API’s sophisticated natural language processing capabilities. By reading and understanding the PDF report, OpenAI can extract key points and summarize the information in a more accessible format.

Using the Python package pymupdf, we scan the PDF and convert its contents to text. This text is then fed to OpenAI API to highlight important sections and summarize the findings, including potential issues and recommended actions.

The Python method process_directory reads each PDF, converts it to text, and then passes this text to another method, generate_summary, which calls the OpenAI API to generate a concise assessment summary.

Method: process_directory()

def process_directory(directory):
    """Processes each PDF file in the given directory to generate a summary."""
    hostname, port_number, database_name = directory.split('_')
    for file in os.listdir(directory):
        if file.endswith('.pdf'):
            file_path = os.path.join(directory, file)
            pdf_text = extract_text_from_pdf(file_path)
            summary = generate_summary(pdf_text)
            print(f"Summary for {file} ({hostname}, {port_number}, {database_name}):\n{summary}\n")

Method: generate_summary()

def generate_summary(text):
    """Generates a summary for the given text using OpenAI's API."""
    response =
            {"role": "system", "content": "You are database \
               reliability engineer providing migration \
               assessment summary."},
            {"role": "user", "content": "Summarise the output \
              of assessment text: \n" + text}
    summary = response.choices[0].message.content.strip()
    return summary

Understanding OpenAI API Parameters

Understanding the role and impact of various OpenAI API parameters is crucial for tailoring your query results. Here’s a brief overview:

temperature (0.4): This parameter controls the level of creativity or randomness in the responses generated by the model. A lower temperature, such as 0.4, results in more predictable and conservative outputs. Conversely, a higher temperature encourages diversity and creativity in the answers.

max_tokens (150): Specifies the maximum length of the generated response measured in tokens (words and characters). Setting this to 150 means the response will not exceed 150 tokens, ensuring concise and to-the-point answers.

top_p (1.0): Also known as “nucleus sampling,” this parameter filters the model’s token generation process. A value of 1.0 means no filtering is applied, allowing any token to be considered. Lowering this value helps in focusing the response generation on more likely token sequences, potentially enhancing relevance and coherence.

frequency_penalty (0.0): Adjusts the likelihood of the model repeating the same line of text. A value of 0.0 implies no penalty on repetition, enabling the model to freely reuse tokens. Increasing this value discourages repetition, fostering more varied and dynamic outputs.

Above python methods generated modest summary from my sandbox environment. Modest at this point – we can take this much further though. I’ve taken small part of whole summary describing migration effort from MS SQL Server 2019 database to RDS for PostgreSQL.

Migration Plan Summary

Source Database

  • AdventureWorks2019.MSSQL
  • Microsoft SQL Server 2019 (RTM-CU22-GDR) – 15.0.4326.1 (X64)
  • Standard Edition (64-bit) on Windows Server 2019 Datacenter
  • Case sensitivity: OFF

Target Platform:

  • AWS RDS for PostgreSQL

Assessment Findings:

  • Storage Objects: 100% can be converted automatically or with minimal changes.
  • Code Objects: 77% can be converted automatically or with minimal changes.
  • Estimated 99.9% of code can be converted to AWS RDS for PostgreSQL automatically.
  • 515 conversion actions recommended ranging from simple to complex tasks

Above AI-generated summary can be a significant time saver for database administrators and engineers. Instead of going through pages of detailed reports, they can quickly glance through the summary and understand the key points. It can also be used as a reference guide during the migration process, helping to avoid potential issues and ensuring a smooth transition.

Building a fully automated OpenAI-Powered Python Module for PDF Analysis and Summary Generation

To generate the assessment summary using the OpenAI API, I developed the Python methods described above. These methods are components of a larger assessment framework that I’m currently developing. In this article, we focus exclusively on the integration with the OpenAI API. It’s worth noting that the PDF files used as input are generated through a fully automated process. However, the details of that process are beyond the scope of this blog post.

Python, with its versatility and powerful capabilities, is ideal for integrating with the OpenAI API. It offers libraries for API interactions and processing PDF files, enabling the automation of the entire workflow—from reading PDF files to generating summaries.

For the initial step, libraries such as PyPDF2, PDFMiner, or pymupdf—which I prefer—can be utilized to read the contents of PDF files. After extracting the text, this information can be processed by the OpenAI API. The API is designed to analyze the text, pinpoint the essential information, and compile a concise summary.

Subsequently, this summary can be saved either as a text file or within a database for easy access in the future. Moreover, the module can be configured to insert summaries into a database table, integrating them into a larger assessment data repository. This data can then be leveraged for generating reports, such as Power BI dashboards or other forms of reporting, allowing key stakeholders to stay informed about the migration process’s progress.

Setting Up Azure OpenAI API: Essentials and Differences from ChatGPT

The Azure OpenAI API is a cloud-based service enabling developers to integrate OpenAI’s capabilities into their applications. To utilize the Azure OpenAI API, one must have an Azure account and subscribe to the OpenAI service, in addition to generating an API key for authentication during API requests.

There are notable differences between utilizing ChatGPT and the Azure OpenAI API.

For ChatGPT, your Python module only requires the openai.api_key to be set, along with specifying the model, such as “gpt-4” in my example code. However, integrating with the Azure OpenAI API necessitates additional configuration:

    openai.api_base = os.getenv('AZURE_OPENAI_ENDPOINT')
    openai.api_key = os.getenv('AZURE_OPENAI_API_KEY')
    openai.api_version = os.getenv('AZURE_OPENAI_VERSION')
    openai.api_type = "azure"
    deployment_id = os.getenv('AZURE_OPENAI_DEPLOYMENTID')

It’s important to note that when using Azure OpenAI, Python OpenAI API parameter model corresponds to your specific deployment name instead of “gpt-4” as it was for ChatGPT model in my examples earlier.

The Azure OpenAI API and ChatGPT OpenAI both offer advanced natural language processing capabilities, albeit tailored to different use cases. The Azure OpenAI API is specifically designed for embedding AI functionalities into applications, whereas ChatGPT OpenAI excels in conversational AI, facilitating human-like text interactions within applications.

Choosing between the two for summarizing database migration assessments hinges on your project’s unique needs. Azure OpenAI API is the preferable option for projects requiring deep AI integration. On the other hand, if your application benefits from conversational AI features, ChatGPT OpenAI is the way to go.

In summary, utilizing the OpenAI API can drastically streamline the database migration assessment process. The AWS Schema Conversion Tool yields a thorough report on your database schema and potential issues, which can efficiently be condensed using the OpenAI API. By developing a Python module, this summarization process becomes automated, thus conserving both time and resources. Regardless of whether Azure OpenAI API or ChatGPT OpenAI is chosen, each offers potent AI capabilities to facilitate your database migration endeavors.

Running benchmark: Comparing Ubuntu 20.04 and RHEL 8.5 performance

The claim

I found a claim on Quora that Ubuntu is slow compared to RHEL. I never thought about it. Is it really? It seemed like a sentimental statement with nothing to prove it. I questioned the claim and found out, that many people tried to support the claim still without providing any kind proof.

Instead of continuing to asks any evidence, I decided to dig the evidence my self.

Is there any difference in the performance between the two? I don’t really know, but if you think about default server install with nothing extra, I doubt there could be any significant performance difference.

Ubuntu 20.04 has currently kernel 5.4 where as RHEL 8.5 has 5.13. Libraries and software are pretty much the same. File system by default is XFS on both. I always enable LVM although that could affect performance – certainly not by improving it, but there are other advantages. I don’t think LVM reduces performance much either and as said I always set it up anyway.

In this case I have two virtual machines both having 4GiB RAM and two 3.3GHz CPU cores running on qemu/kvm. The host OS is, yes you guessed right, Ubuntu 20.04 , because on desktop it has certain software I need. And, it works well as virtual host too. That does not affect the results anyhow, since the guest OS has no idea of host OS.

I haven’t tuned either of the guests OS at all except for one thing. I set tuned profile to virtual-guest for both, which makes sense. It is the recommended profile when I run tuned-adm recommend on both of the guests machines.

Put the HammerDB down

Last I ran HammerDB I had to settle with text based version, but this time it had a nice working GUI. But even before quick HammerDB installation, I downloaded Db2 11.5.7 Community Edition. Installed it on both Ubuntu and RHEL. I created SAMPLE database with db2sampl and took timing for that: no difference really. I knew it. Ok that doesn’t prove anything.

But, the real test does. HammerDB.

Ubuntu 20.04

Let’s start with Ubuntu. I read a tutorial on how to run time based benchmark with HammerDB. I want to do this fast. One virtual user only. Looks good.

Ubuntu 20.04

The process goes:

  • Choose Engine and configure it (Db2)
  • Build schema
  • Configure and load driver
  • Configure virtual user
  • Create virtual user(s)
  • Run virtual users
  • Monitor and wait
Here are the results for Ubuntu.

The results for Ubuntu 20.04

System achieved 5178 NPM from 22865 Db2 TPM

RHEL 8.5

Then same thing for RHEL. The machine crashes twice. Reminds me of kernel parameters. We have only 4GiB, so might be I need to tune them. But no, it run all good the third time. In my previous job though, servers with low memory running Db2 on RHEL crashed always without tuning the kernel parameters. There’s a simple formula based on RAM to calculate correct values for Db2 here. That said, I did not change anything from defaults for Ubuntu nor RHEL. It wouldn’t be fair comparison, if I started to tune the kernel parameters for one and not to the other.

Running on RHEL 8.5
There’s finally some I/O wait
The winner is RHEL 8.5 by two New Orders Per Minute (NOPM)

The results for RHEL 8.5

System achieved 5180 NOPM from 22815 Db2 Db2 TPM

First conclusion

There is really no difference between Ubuntu and RHEL what comes to achieved performance results. The two new orders per minute makes 0.04% difference which I’m pretty sure no one can notice just by “using the server a bit”.

Comparison between database engines

Since I already started playing with HammerDB, why not try some more tests. I have earlier installed Db2 on the host machine itself as well as MS SQL Server. I also have virtual machine running Oracle Linux 8 on it with the same 4GiB RAM and two CPU core setup. MySQL and PostgreSQL I have running on the host itself.

The hosts OS, as said, is running Ubuntu Desktop 20.04. It has 4 x 3.3GHz cores and 32GiB RAM and fast NVMe 500GiB M.2 PCIe SSD. This is small form factor machine suitable for industrial use as a headless server running for example Linux. Or you can use it as desktop computer as well. My idea for it was to use it as a platform for several virtual guests, but I wanted to see how it works as a Linux desktop computer as well.

Let’s do few quick tests on the host itself for various database engines. More of a test of HammerDB itself than real comparison between the engines.


TEST RESULT Ubuntu 20.04 Desktop: System Achieved 6651 NOPM from 2928 TPM.

I’m a bit surprised it didn’t achieve more. Need to test more. It takes time for bufferpools to warm up with automatic memory tuning and with 32GiB memory I’m pretty sure we could get much better results.

MS SQL Server

But let’s check with MS SQL Server I have running on the same machine. Certainly Db2 beat MS SQL Server, right?

MS SQL Server gets higher TPM numbers compared to virtual machines
Obviously the benchmark is somewhat different between the engines,

The winner is… oh no, MS SQL Server

8394 New Orders Per Minute with 19243 SQL Server TPM


I have one Oracle 21c Server running on VM running Oracle Linux 8. Oh but Oracle – I’m so lost with it. HammerDB asks too much questions and it seems I need to create another pluggable database. I will do that – later.

PostgeSQL and MySQL

Out of curiosity I ran the test for PostgreSQL and MYSQL:

PostgreSQL: TEST RESULT: 11607 NOPM from 26880 PostgeSQL TPM


I have no idea why the difference between above two is that significant. Might be for various reasons. I wouldn’t pay much attention on the difference since running the test on host OS and not on virtual machines with proper setup doesn’t make much sense – unlike the more serious comparison I did for Ubuntu Server and RHEL.

Final Conclusion

Without official test for Oracle we cannot make any other conclusion than Oracle is the slowest from these three DB Engines: Db2, MS SQL Server and Oracle. I’m kidding of course; I’m no Oracle expert and just too slow myself to set a proper test for Oracle. That might change once I have enough time to dig deeper on Oracle. For MySQL and PostgreSQL the test was also too quick; more of a test do they work similarly in comparison to Db2 and MS SQL what comes to HammerDB.

What comes to the original claim about Ubuntu being overall slow and which surprisingly many is willing to believe, I think I have busted the claim.

We can speculate how about real server environments and please do, but before you actually have any benchmarks to show otherwise, I take it proven that Ubuntu and RHEL are equally slow or fast.

Also, what comes to MS SQL Server performance compared to Db2, obviously this was not the last word. Let’s try with 10 virtual users beating Db2 for a bit longer.

Db2 with 10 Virtual Users

Final results for Db2 running on this tiny Asus Mini PC PN41 were:

TEST RESULT: System achieved 15650 NOPM from 68735 Db2 TPM.

So we have a winner: Db2 11.5.7?

In a sense Db2 won that it did get the highest number of new orders per minute yes. But in comparing with other database engines I didn’t really organise any meaningful tests between them this time.

Want to test yourself?

Prove me wrong. Run your own tests and provide me your data and conclusions. I have serious doubt Ubuntu Server and RHEL differs much what comes to performance. There certainly is plenty of other things which makes the difference when choosing the distribution. Things like support, cost, platform you are running on and so on. Red Hat certainly has it’s advantages on enterprise level support whereas Ubuntu started strong on desktop, but it is easy to deploy for example on Azure and fully supported.

Configure SQuirreL SQL Client for DB2 for Linux, UNIX and Windows

Configure SQuirreL SQL Client for DB2 for Linux UNIX, and Windows

At least for me the default settings for SQuirreL DB2 driver didn’t work. That’s why I documented here how to modify the default “IBM DB2 App Driver” so it works.

Configure the DB2 driver

  1. First I copied db2jcc4.jar from the DB2 instance. In my case the driver was under

    /opt/ibm/db2/V11.1/java/ but you find it as DB2 instance user under $HOME/sqllib/java as well, since this is symlink to installation directory.

  2. I stored DB2 db2jcc4.jar  under $HOME/Java, but you can choose a different location.
  3. Click Drivers tab and then double click “IBM DB2 App Driver” to modify it.
  4. Highlight “Extra Class Path” tab and click “Add” to add the db2jcc4.jar you have earlier stored to you computer.
  5. Set “Example URL” to jdbc:db2://<HOST>:<PORT>/<DATABASE_NAME>
  6. Set “Class Name” to

Setup connection to DB2 database

  1. To set up your first DB2 database connection on SQuirreL SQL Client, select the “Aliases” tab and click  icon, to add new database connection.
  2. From “Driver pull down list select “IBM DB2 App Driver”.
  3. Choose “Name” for you connection ie. the database name you are connecting to.
  4. URL is the form you set up when creating the driver: “jdbc:db2://<HOST>:<PORT>/<DATABASE_NAME>”. Set “host” and “dbname” accordingly.

No you can test your connection buy hitting test and then connect:

Ready to go

Once connection is done you can connect to DB2 database:

And start querying:


If you have any questions or feedback, please connect with me.

Create Linux VM running CentOS 7.3 minimal with pyodbc and Netezza Client

This document describes how to create Linux Virtual Machine (VM) to be run on macOS or Windows Host. When followed the steps in this document, you will have CentOS 7.3 VM capable of running Netezza Linux Client, unixODBC, Python 3.6 with pyodbc and pandas among others. This setup is useful for developing Python code which needs Netezza connection.

Especially macOS users will benefit from this kind of setup, since there is no Netezza client for macOS.

This document concentrates on deploying the VM on VirtualBox, but the CentOS setup portion is identical also when using other hypervisors ie. VMWare Player, VMWare Workstation or VMWare Fusion.

Note: The LinuxVM created in this documented has all capabilities on Python 3.6. You execute python code calling python3.6 instead of just python, which points to python 2.75.

Install CentOS 7.3 on VirtualBox

  1. Download newest version of VirtualBox and install it:
  2. Once installed go to VirtualBox menu and choose “Preferences” and click “Network”.
  3. Click “Host-only Networks” and choose icon add: 
  4. Now you have new “Host-only Network” which is needed for incoming connections. You can check the details by double clicking vboxnet0:
  5. Next create VM with two network adapters:
    1. Choose “New” and select “Name”, “Type” and “Version”:
    2. Click continue. You can keep the memory on 1024MB which is the default.
    3. Click continue and choose “Create virtual hard disk now” and click “Create”.
    4.  “Hard disk type” can be VDI, if you do not plan to run VM on other hypervisors, but if you plan to run it on VMWare hypervisor, choose VMDK. Click “Continue”.
    5. For flexibility choose “Dynamically allocated” and for best performance choose “Fixed size”.
    6. For most purposes 8.0GB is enough, but your needs may vary. Choose “Create”.
    7. Now VM is created, but we need to change some of the network settings:
      1. While new VM is highlighted, choose “Settings” and select “Network” tab.
      2. “Adapter 1” default settings are ok for most cases, but we need to add “Adapter 2” so click “Adapter 2”. We need 2nd Network card for incoming connections, so we select “Enable Network Adapter” and set “Attached to: Host-only Adapter”:
      3. Click “OK”.
  6. We need to have CentOS minimal installation image which we can download from CentOS site:
  7. Choose your download site and store the image to desired location. We need it only during intallation.
  8. Once downloaded, go back to VM Settings on virtual box:
    1. Select “Storage” tab and “Controller: IDE” and click the CDROM icon and then another CDROM icon on right side from “Optical Drive” selection and “Choose Virtual Optical Disk File”.
    2. Select the CentOS minimal installation disk image you downloaded on previous step:

      1. Click “OK”
  9. Now we can start the CentOS minimal installation. Choose “Install….” when VM has booted.
  10. Next we get graphical installation screen. We can keep language settings as default and click “Continue”:
  11. Note when you click the VM, it will grab the mouse. To release the mouse, click left CMD (on MacOS).
  12. Click “Network & Host name”. You can specify your hostname as preferred.
  13. Both Network cards are off by default. Set them both to “ON”.
  14. For both Network cards, click “Configure” and on “General” tab choose “Automatically connect to this network when it is available”:
  15. Click “Done” to get out from Network settings, and click “Installation Destination” to confirm storage device selected by default is correct (no need to change anything). Then click “Done” to get back to main screen and you can start the installation by selecting “Begin Installation”.
  16. During installation set root password and select to create user. In my examples for setting up Netezza client, I have chosen to create “Netezza User” with username “nz”. I will also make this user an administrator:
  17. Once installation is done you can click “Finnish configuration” and then “Reboot”.
  18. VM boots now first time. You can either ssh to the system (from Terminal on MacOS, or using Putty on Windows).
  19. If this is only VM using Host-only network on VirtualBox, it’s likely the IP is You can check the IP for device enp0S8 with command: ip addr show when logged in through VirtualBox console.
  20. After installation first thing to do is to update all packages with yum update command. Either as root give command “yum -y update” or as administrative user as “sudo yum -y update”.

Configure file sharing between Host and Guest OS

You might want to be able to share, for instance your PycharmProjects folder to run Python code you developed directly on LinuxVM. That is a bit of the whole point for the LinuxVM in this case.

To achieve that, you need to enable file sharing. There is few additional steps needed I’l go through below:

  1. You need few additional packages first. Run following commands as root:
    1. yum -y update
    2. yum -y install gcc kernel-devel make bzip2
    3. reboot
  2. Once LinuxVM has rebooted and you have LinuxVM Window active select from menu “Devices” –> “Insert Guest Additions CD Image…” . Then log in to LinuxVM as root via VirtualBox console or ssh again and run following commands:
    1. mkdir /cdrom
    2. mount /dev/cdrom /cdrom
    3. /cdrom/
  3. Now select the folder you want to share from your Host OS to LinuxVM. Go to VM settings and choose “Shared Folders” tab and click icon and then choose the folder you want to share:
  4. Note: Make sure you set the mount permanent. No need for automount option, since we do it a bit differently below.
  5. Above we are sharing PycharmProjects folder. We want to have PycharmProjects folder mounted on LinuxVM on nz users home directory. As nz user we first create directory PycharmProjects with command: mkdir $HOME/PycharmProjects
  6. Then as root, we add following entry to /etc/fstab:PycharmProjects /home/nz/PycharmProjects vboxsf uid=nz,gid=nz           0 0
  7. After reboot you should now have your PycharProjects folder mounted with read and write access under nz users home directory.

    Note: The purpose for above share is, that when you develop your Python code with Pycharm on MacOS and if your code needs connection to Netezza, you can not run it on MacOS, since there is no Netezza drivers. Instead, when following this guide, you will be able to run you Pycharm edited code seamlessly on the LinuxVM through ssh connection, and once confirmed to work, you can commit your changes.

Install Python 3.6 with pyodbc, pandas and sqlalchemy

Log in to LinuxVM as root and run following commands:

yum -y update
yum -y install git
yum -y install yum-utils
yum -y groupinstall development
yum -y install
yum -y install python36u
yum -y install python36u-pip
yum -y install python36u-devel
pip3.6 install pandas
yum -y install unixODBC-devel
pip3.6 install pyodbc
yum -y install gcc-c++
yum -y install python-devel
yum -y install telnet
yum -y install compat-libstdc++-33.i686
yum -y install zlib-1.2.7-17.el7.i686
yum -y install ncurses-libs-5.9-13.20130511.el7.i686
yum -y install libcom_err-1.42.9-9.el7.i686
yum -y install wget
yum -y install net-tools
pip3.6 install sqlalchemy
pip3.6 install psycopg2

Testing pyodbc

Edit the connection string accordingly:

[nz@nzlinux ~]$ python3.6
Python 3.6.2 (default, Jul 18 2017, 22:59:34) 
[GCC 4.8.5 20150623 (Red Hat 4.8.5-11)] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import pyodbc
>>> pyodbc.connect(server="nz", database="TEST", dsn="NZSQL", user="admin", PWD="password", autocommit=False)
<pyodbc.Connection object at 0x7f66ef8566b0>

Install Netezza Linux client

First you need to download NPS Linux client from IBM Fix Central

Then, as root run following commands (accept all defaults):

mkdir NPS
cd NPS
tar xvfz ../nz-linuxclient-v7.2.1.4-P2.tar.gz
cd linux
cd ../linux64

Now, log in as nz user and add following lines to $HOME/.bashrc (modify credentials and server details accordingly: NZ_USER, NZ_PASSWORD and NZ_HOST):
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/nz/lib64
export PATH=$PATH:/usr/local/nz/bin
export ODBCINI=$HOME/.odbc.ini

To make above changes effective without logging out and in, you can instead run command: . ./.bashrc

Now you should be able to use nzsql:

[nz@nzlinux ~]$ nzsql
Welcome to nzsql, the IBM Netezza SQL interactive terminal.
Type:  \h for help with SQL commands
       \? for help on internal slash commands
       \g or terminate with semicolon to execute query
       \q to quit

Setup ODBC

Copy following two files, odbc.ini and odbcinst.ini to /etc as root:


As nz user create following symlinks:

ln -s /etc/odbcinst.ini .
ln -s /etc/odbc.ini .
ln -s /etc/odbc.ini .odbc.ini
ln -s /etc/odbcinst.ini .odbcinst.ini


If you have any questions, please connect with me.


The .odbc.ini and .odbcinst.ini issues seems to be fixed with newer Python versions, so creating symlinks to users home directory nor creating system files under /etc are not anymore required. Just using .odbc.ini and .odbcinst.ini in user’s home directory works now as it is supposed to work.

How to set up SQuirreL SQL Client for Netezza

This article describes how to set up SQuirreL SQL Client for Netezza. There is not much free SQL query tools available for MacOS and Linux, but SQuirreL is an exception. It uses JDBC for connecting to Netezza, so first thing you need is Netezza JDBC driver. Netezza JDBC driver you can find  for example from latest Netezza Linux client, for example (it is inside npsclient.7.2.1.X-PX.tar.gz  as lib/nzjdbc3.jar).

Download and install SQuirreL SQL Client

  1. Download SQuirreL jar installer package from:
  2. You need to have Java JDK installed to run and install SQuirreL. As of writing this HOWTO latest JDK can be downloaded from:
  3. Once the jar installer is downloaded, run it with: java -jar squirrel-sql-3.7.1-MACOSX-install.jar
  4. Accept the defaults and choose which plugins you need. In my example I have chosen plugins “Multi source”, “Data import”, “MySQL”, “Netezza”, “Oracle”, “PostgreSQL”, “Session Scripts”. “Smart Tools”, “SQL Parametrisation”, “SQL Replace” and “SQL Validator” .
  5. Click Next and Done and you have SQuirreL SQL Client installed on you system.

Configure SQuirreL SQL Client for Netezza

Before beginning configuring the driver, you need to have Netezza JDBC driver nzjdbc3.jar stored on you computer. Netezza JDBC driver can be found for example from Netezza Linux Client under the 32bit Linux tar archive.

Configure the Netezza driver

Once SQuirreL SQL Client is started, first thing to do, is to add Netezza JDBC driver to available drivers on SQuirrelL SQL CLient. In Netezza’s case, there is no Netezza driver on list of drivers, so we will add one.

  1. I stored Netezza JDBC driver under $HOME/Java, but you can choose a different location.
  2. Click Drivers tab and and then click  sign to add new driver.
  3. Highlight “Extra Class Path” tab and click “Add” to add the Netezza JDBC driver you have earlier stored to you computer.
  4. Set “Name” to “Netezza”, “Example URL” to “jdbc:netezza://<host>:5480/<dbname>”, “Websitete URL” to “” and most importantly set “Class Name” to “org.netezza.Driver”. Once done, click “OK”.

Setup connection to Netezza database

Since SQuirreL SQL Client uses JDBC for connecting to database, it can in theory have only one connection per database. So you need to create one connection per database you are working on because of the JDBC limitation. There is commercial tools like RazorSQL which you can have only one database connection and still connect and query all databases with that same connection, but with SQuirreL SQL Client this is not the case.

  1. To set up your first Netezza database connection on SQuirreL SQL Client, select the “Aliases” tab and click  icon, to add new database connection.
  2. From “Driver pull down list select “Netezza”.
  3. Choose “Name” for you connection ie. the database name you are connecting to.
  4. URL is the form you set up when creating the driver: “jdbc:netezza://<host>:5480/<dbname>”. Set “host” and “dbname” accordingly.
  5. Now you can test the connection. Click “Test” and then “Connect”. If connection is unsuccesful, check the trace to find out the reason.
  6. Once alias is created, you can connect to database. Either double click the alaias or right click it and select connect.


If you have any questions or feedback, please connect with me.

Five steps to tune Netezza query performance

Netezza is designed with simplicity in mind. You can get it up and running in hours rather than weeks. When you follow the basic rules, 99 percent of your applications and queries can perform well. There are six things you should keep in mind while designing your databases and setting up maintenance tasks:

  • Distribution
  • Data types
  • Statistics
  • Zone maps
  • Data organization
  • Groom

When you have taken care of the these things, you shouldn’t have any major issues with performance. However, if you do have issues, how do you find them? Follow these five steps, which can help you to find and possibly fix the performance issues on your appliance.

1. Use IBM Netezza Performance Portal and the query history database

The IBM Netezza Performance Portal is an excellent tool for making sure you have everything in place, and if you don’t it will help you to identify any issues. It provides an excellent front end to the query history database and it is able to connect the appliance performance history with your query history.

Netezza Performance Portal and its installation guide are available for download from IBM Fix Central. The installation is fairly simple. Please refer to the “IBM Netezza Performance Portal User’s Guide” (which is included in the download) for installing and configuring both Netezza Performance Portal and the query history database.

2. Check the server load and resources

One of the greatest things about Netezza Performance Portal is its ability to monitor one or more Netezza systems and their resource usage. From the nice graphical user interface (GUI) you can easily identify performance peaks. Then, if you see any changes in trend, you can drill down and take a closer look by using your mouse pointer to click where the change begins and ends. You can repeat this as many times as you want. After zooming in on the period of time you are interested in, you can click the “Jump to History” button to see which queries were running during that time slot. But first you will need to choose the host you are interested in, as this will populate the “Submit Time” and “Finish Time” fields in the query history view.

3. Identify the problematic queries

Identifying problematic queries is, of course, easier said than done. Basically, there are three different types of long-lasting queries that usually require a closer look:

  • Queries that take a long time to finish because they need to access a lot of data.
  • Queries that take long time to finish because the query is not optimal.
  • Queries that take a long time to finish because the database design is not optimal.

When you first look at your queries, they probably just look like long-lasting queries. However, if you know your data and your queries, you might be able to place at least some of them in the first class I mentioned (queries that naturally take a long time to finish because of large amounts of data). What I usually do myself, after I have zoomed in on a performance peak or otherwise interesting period from the Netezza Performance Portal monitoring view, is sort the queries based on their “Query Duration.” I simply list the queries in descending order and then look at the “Query Text.”

When you have selected the interesting query, you can perform various actions with Netezza Performance Portal, including the following:

  • You can right-click the query and check the “Identical Query Trend Chart.” This will give you an idea of the variation in duration for identical queries over time. For instance, if the system is overwhelmed by concurrently running workloads, it is obvious that a query will not run as fast as it normally would. If you notice that a query took longer than normal to run, you should check what else was running on the system at that time.
  • If it really took longer to run than is typical, you can check the “Query/Plan Activity Chart.” This will give you a nice graphical view of all the queries running concurrently on the system, which could be affecting the duration of the query you were interested in.
  • You can check if the statistics are up to date on tables related to the query, and you can even update the statistics thorough Netezza Performance Portal if they are outdated.You can also check encumbrance. There might have been, for instance, loads or aborted ad hoc queries running on the system that negatively affected the system performance. I have once identified the latter to be the case for why highly-prioritized extract, transform and load (ETL) tasks did not finish in time. Since the queries were aborted, they were not seen in the Query/Plan Activity Chart, but rather on the encumbrance view.

Picture 1: Identical Query Trend Chart

4. Check the query plan

You can check the query plan directly from Netezza Performance Portal and that’s not a bad choice. However, what I usually do is check only the plan ID. You can find this from one of the columns when you are listing the queries on the query history view. I then take the plan ID, log in to the appliance and use nz_plan to take a closer look at the query. One advantage of nz_plan is that it lists the interesting snippets early in the file. Another attractive feature is that it rewrites the query very nicely in a readable format.

That said, you can still use the other techniques available to produce the query plan, including the one available directly through Netezza Performance Portal.

Picture 2: Query Plan generated with nz_plan


5. Check the distribution keys and change them if needed

Now that you have the query plan, one of the first things you can check from there are the distribution keys for the tables the query is accessing. Are they what you assumed they were? Are there re-distributions—single or double? If you see something like “1[03]:spu DownloadTableNode distribute into link 2147484337” you know that the table is distributed. If you assumed it isn’t, then you should check the distribution keys again.

Check how it works now when there are no issues

Don’t wait until you have issues. Familiarize yourself on how to monitor performance issues before you have performance issues. If you don’t already have Netezza Performance Portal, install and deploy it. Try and test how nz_plan utility works. Read the query plans. By doing this, you will be ahead of the game and ready to tackle any upcoming issues.

If you have any questions or suggestions related to query performance tuning, please leave a comment. You can also follow me on Twitter @TVaattanen to discuss more about Netezza.

Everything you wanted to know about networking but were afraid to ask (Part Three)

This blog post is the third part of a series about questions you may have wanted to ask about Netezza networking. The first part concentrated on basic Netezza networking, while the second part continued with network bonding and floating IP addresses. This is the third part, which concentrates on advanced configuration options.

Network speed

By default, a Netezza appliance host has two available Peripheral Component Interconnect (PCI) slots for additional PCI cards. Normally you would use one for a 10 GB dual port Network Interface Adapter (NIC) and the second available slot for dual port 8 GB Host Bus Adapters (HBA). The first you could use for 10 GB networking, and the second could be used for Storage Area Networking (SAN) or LAN-Free backups.

Internally, the appliance uses 10 GB networking. Externally, the default is 1 GB. If you want to have 10 GB external networking, then you need to have the additional 10 GB dual port NIC. Assuming you have a 10 GB network infrastructure in place, you most probably want to go directly to 10 GB.

Even if you plan to initially start with 1 GB external networking, you should consider getting the additional 10 GB NIC and 8 GB Host Bus Adapter (HBA), because you are likely going to use them later.

More about network bonding

By default, the appliance has two hosts. Both of the hosts have one external bonded virtual network device, which consists of two physical 1 GB network interfaces. By default, the network bond is created as active/passive, so the maximum bandwidth you can achieve is 1 GB. If you ask, and your network switch supports link aggregation, you can configure the network bond as Active/Active to get a 2 GB link.

As mentioned above, there are two available PCI slots. This means you can also add two 10 GB dual port NICs to those slots. That way, you can bond up to four 10 GB physical network devices together to achieve maximum 40 GB bandwidth.

Another option would be to use two of the 10 GB ports for virtual IP addresses for application connectivity, and the two remaining ones for a backup network. There are plenty of options, when you consider that you can bond together any of the 10 GB ports in any order to create a bonded device, and then you can choose to go for active/active or active/passive mode.

What about LAN-Free?

This section doesn’t actually cover pure TCP/IP networking, but rather connectivity without TCP/IP. As mentioned earlier, you can have 8 GB HBA installed on one or both of the available PCI slots on the hosts. If you decide to have at least one available PCI slot for additional 8 GB HBA, you could use it for LAN-Free backups.

TCP/IP networking is usually done in shared mode, so you have to share the bandwidth with other users—unless you have a dedicated link, which most often you don’t have. With SAN it is easier and more common to create a dedicated link between the appliance and, for example, the backup server. Or you can connect to an external SAN disk through a dedicated link. That of course has clear benefits; when you know exactly how much bandwidth there is and when you don’t need to share it with anyone.

Another benefit with the LAN-Free option is is the CPU usage. TCP/IP implementations tend to have more CPU overhead compared to SAN. I would emphasize the benefit of the dedicated link though, since CPU on the host is rarely limited while dealing with backups, for instance.

Management interfaces

I already mentioned the management IP addresses: usually two per host, one being the host IP itself, and the other being the IP address of the integrated management module (IMM).

The IMM IP addresses are extremely handy if the host itself is not reachable through the host IP due to the fact it has failed with a hardware error, or if there is something wrong with the configuration. Through IMM, you get console access though the web interface, and either debug the problem or fix the configuration issue.

Some clients require a separate management IP, which is not attached to any network devices used by applications and which still has direct TCP/IP connectivity to the host. In this case neither the host IP nor the IMM IP can be used; you need to use some other available physical network port or interface. If this is the case, you should clearly define the requirements, so you can check the available options.

What else?

If anything else is on your mind that you did not dare to ask earlier, feel free to ask or comment below. You can also follow me on Twitter @TVaattanen to discuss more about Netezza.

Everything you wanted to know about Netezza networking but were afraid to ask (Part Two)

This blog post is the second in a three-part series with the goal of answering questions you might have about Netezza networking. The first part concentrates on basic Netezza networking, whereas this second part covers more advanced networking concepts. For advanced configuration options, you can check out the upcoming third part of this blog post.

Network bonding

You have two hosts: active and passive. Each has its own IP address. These IP addresses are not floating. These are called host IPs. Since you want to have maximum redundancy on all components, there are actually two physical network devices virtually bound together to create virtual networking devices (one for each host). Both hosts have two physical network devices that carry one IP address. This is called network bonding.

Let’s say both of the hosts have network devices eth6 and eth7 and they create a coupled virtual device called bond2. We usually use bond0 and bond1 internally, so the first bonded device for external use is normally bond2.

For the virtual device bond2, you can assign an IP address and connect to a host. Both active and passive hosts will have this device and both of the hosts will have their own individual IP address, which is bound to this virtual device.

Virtual IP

If you think of this from an applications point of view, it wouldn’t make sense to connect to the host IP, since if the active host fails, you would need to re-configure applications to use the new active host, which has a different IP.

That’s why applications use virtual IP. Virtual IP is actually an IP alias, which is bound to an active host. Hosts run standard Linux operating systems, so if you are familiar with Linux, it’s easy to explain. If not, it’s still not rocket science. On Linux, you can easily add IP aliases on top of any physical, or virtual for that matter, network device . If you have physical network device eth0 with fictional IP address, you can add another IP address to that same physical device just by assigning an IP to device eth0:0. Next you add to device eth0:1 and so on.

In this case, you have virtual network device bond2, which is a bonded device having physical devices eth6 and eth7 behind it. If you lose eth6, you are still good as long as physical device eth7 is good. To connect to either of the hosts directly, you would use the IP address assigned to bond2 on the particular host, or rather the host name you have assigned in your domain name server (DNS) for that IP address.

Floating IP

As I said, applications connect to a virtual IP. The virtual IP is assigned to virtual network device bond2:0. It only exists on an active host. This is something called a floating IP, and it is always on the active host. If Host 1 fails, it will be on Host 2. If, as in my example, device eth6 fails, you have bonded device bond2, which consists of eth6 and eth7, the floating IP is still good on that same appliance as before.

There are two virtualization layers here. One is done though network bonding, the other is done through cluster software. If one of the network devices physically breaks, the network bonding will do the trick, and you are still good to go. If the other appliance breaks, you have clustering software, which can deactivate the bond2:0 on the failing host and create bond2:0 on new active host.

So the bond2:0 always has the virtual IP your applications are able to use. You should, of course, always assign host names in your DNS for this virtual IP, and use this host name in your applications instead of using IP addresses directly. That way, if you ever need to change the IP address for the virtual IP, you don’t need to change configurations for several applications. Instead, you just have to change the IP for the host name you have defined for the virtual IP in your DNS configuration.

What about changes to the default configuration?

I will cover advanced configuration options in part three of this blog post. If you have any network-related questions or suggestions, please add them below in the comments. You can also follow me on Twitter @TVaattanen to discuss more about Netezza.