Category Archives: Linux

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:

  • DATETIME to TIMESTAMP: PostgreSQL uses TIMESTAMP instead of DATETIME.
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 USER ha WITH ENCRYPTED PASSWORD 'yourpassword';
GRANT ALL PRIVILEGES ON DATABASE homeassistant TO ha;

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

Instructions:

  • 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:

recorder:
  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.

Conclusion

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

Introduction

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.

Conclusions

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.

References

Five easy steps for setting up SSL on HomeAssistant utilising Let’s Encrypt Certbot

Securing your HomeAssistant setup should be a priority, especially if you plan on accessing your system remotely. One of the best ways to do this is by setting up an SSL certificate. This article guides you through five easy steps to set up SSL on HomeAssistant using Let’s Encrypt Certbot.

Understanding the Importance of SSL for HomeAssistant

Secure Sockets Layer, popularly known as SSL, is a security protocol that encrypts the connection between a web server and a client. When implemented on your HomeAssistant, it prevents eavesdropping and tampering of your data by encrypting all communication between your HomeAssistant and your devices. This is crucial, especially when accessing your HomeAssistant remotely over the internet where your data could be intercepted.

Moreover, SSL also provides authentication, ensuring that you’re communicating with the right server and not a malicious one. This is achieved through the use of SSL certificates issued by trusted Certificate Authorities (CAs). These certificates also provide visual cues, such as a padlock symbol, giving end-users confidence that their connection is secure.

An Overview of Let’s Encrypt Certbot

Let’s Encrypt is a free, automated, and open Certificate Authority. It provides digital certificates needed to enable HTTPS (SSL/TLS) for websites. The Certbot is an easy-to-use client that fetches certificates from Let’s Encrypt and configures your web server to use them.

By using Let’s Encrypt Certbot, you can easily acquire and renew SSL certificates for your HomeAssistant. It automates the process of obtaining and installing SSL certificates, thereby saving time and eliminating the risk of manual errors. Moreover, it also handles the renewal of SSL certificates, ensuring that your connection remains secure.

Contrary to what seems to be the case for many, if not most, I find the use of third-party VPN solutions for accessing an otherwise cloud-free HomeAssistant setup to be illogical. Moreover, the notion of implementing the HomeAssistant Cloud service, Nabucasa, doesn’t appeal to me at all. The core of my philosophy is to maintain a smart home solution that is independent of both third-party and cloud services.

Step 1: Installing Let’s Encrypt Certbot

The initial step to enable SSL for your HomeAssistant involves installing Let’s Encrypt’s Certbot. The installation method differs across operating systems. On Linux systems, it’s straightforward to install Certbot using the package manager. For example, Ubuntu users can execute the command sudo apt-get install certbot.

My setup took a slightly different route. As previously mentioned, my HomeAssistant operates within a Docker container, and I also host several websites, including the one hosting this blog post, on a virtual machine. This VM shares the same server as the HomeAssistant Docker container. Installing Certbot on CentOS Stream, the operating system of my VM where SSL is primarily needed, was a breeze by simply following the guided instructions available on the Certbot website.

You can confirm the successful installation of Certbot by executing certbot --version in your terminal. This command should return the version number of Certbot installed on your machine. Should you encounter any issues, indicating that Certbot hasn’t been installed properly, you may need to address the installation process or attempt reinstalling it.

Step 2: Generating an SSL Certificate

With Certbot installed, the subsequent step involves generating an SSL certificate for your domains. In my experience, executing the command certbot --apache was a straightforward process. Certbot intelligently scanned all my Apache virtual hosts, generating certificates for each. Interestingly, it selected the first domain in the list as the root certificate for all others—a decision I wouldn’t have made intentionally, but one I’m content with nonetheless.

Aiming to secure a certificate for HomeAssistant as well, I introduced fake virtual hosts within Apache and initiated certbot --apache once more, this time specifying the addition of the exclusive HomeAssistant domain, which for me is ha.auroranrunner.com.

An alternative method involves the command certbot certonly --standalone. This approach instructs Certbot to secure a certificate by functioning as a temporary web server (standalone) to authenticate domain ownership—useful for situations requiring a more hands-off approach.

However, my objective was for Certbot to manage the certification updates for all domains collectively, thus I adopted a slightly different strategy.

Opting to exclusively focus on HomeAssistant, without intertwining Apache configurations, prompts a straightforward process. You’ll be asked to input your domain name along with your contact details. Upon submission, Certbot seamlessly liaises with the Let’s Encrypt Certificate Authority (CA), generating an SSL certificate for your domain. The newly minted certificate and its private key are securely stored in the directory /etc/letsencrypt/live/your_domain_name/.

Step 3: Setting SSL sync between primary host and secondary host

In my situation, it was necessary to establish a method for synchronizing the SSL certificates between the virtual machine hosting the Apache web servers and the server operating the HomeAssistant Docker container. To accomplish this, I undertook the following steps:

  1. Established passwordless SSH authentication between my Apache hosts and the server hosting HomeAssistant to ensure a seamless connection.
  2. Created a script located at /usr/local/bin/sync_lets_cert designed to facilitate the synchronization of Let’s Encrypt certificates.
  3. Developed a systemd service aimed at automating the daily synchronization of Let’s Encrypt certificates between the two hosts, ensuring that both systems always use the latest SSL certificates.
  4. Configured a dedicated volume for the HomeAssistant Docker container mapped to /etc/letsencrypt:/etc/letsencrypt. This setup allows the HomeAssistant container direct access to the synchronized SSL certificates, simplifying the process of securing communications.

The script located at /usr/local/bin/sync_lets_cert is responsible for synchronizing the SSL certificates between servers. Its contents are as follows:

#!/bin/bash

# Variables
SECONDARY_SERVER="my_vm_host_server"
DOMAIN="ha.auroranrunner.com"
LIVE_PATH="/etc/letsencrypt/live/$DOMAIN"
ARCHIVE_PATH="/etc/letsencrypt/archive/$DOMAIN"
DEST_LIVE_PATH="/etc/letsencrypt/live/$DOMAIN"
DEST_ARCHIVE_PATH="/etc/letsencrypt/archive/$DOMAIN"

# Sync the live directory
rsync -avz -e ssh $LIVE_PATH/ $SECONDARY_SERVER:$DEST_LIVE_PATH

# Sync the archive directory
rsync -avz -e ssh $ARCHIVE_PATH/ $SECONDARY_SERVER:$DEST_ARCHIVE_PATH

This script ensures that the certification files are kept in sync between the hosts. The next step involves setting up a systemd service to schedule this script’s execution, which proved to be slightly more complex but was successfully achieved as follows:

  1. Create a timer file at /etc/systemd/system/sync_lets_cert.timer with the following content to establish a daily execution schedule:
[Unit]
Description=Daily timer for Let's Encrypt certificate sync

[Timer]
OnCalendar=daily
Persistent=true

[Install]
WantedBy=timers.target
  1. Then, create the service file /etc/systemd/system/sync_lets_cert.service to define the synchronization task:
[Unit]
Description=Sync Let's Encrypt Certificates

[Service]
Type=oneshot
ExecStart=/usr/local/bin/sync_lets_cert
  1. Finally, start and enable the service and timer with the following commands:
systemctl start sync_lets_cert.service
systemctl enable sync_lets_cert.timer

With these steps completed, the SSL certificates will not only be renewed every 90 days but also synchronized between servers daily, ensuring seamless security and authentication continuity.

Step 4: Setting up SSL on HomeAssistant

With the SSL certificate secured, the following step is to integrate SSL into your HomeAssistant setup. This process entails adjusting your HomeAssistant’s configuration to recognize and utilize the SSL certificate. Achieve this by appending the below entries into your HomeAssistant’s configuration.yaml file:

http:
  ssl_certificate: /etc/letsencrypt/live/ha.auroranrunner.com/fullchain.pem
  ssl_key: /etc/letsencrypt/live/ha.auroranrunner.com/privkey.pem
  base_url: https://ha.auroranrunner.com:8123

These lines instruct HomeAssistant on the locations of the SSL certificate (fullchain.pem) and its corresponding private key (privkey.pem). Post addition, a restart of your HomeAssistant is required for the adjustments to be applied.

Initially, setting up SSL without specifying base_url sufficed for web browser access. However, to ensure the mobile application functioned correctly, including the base_url became necessary.

Regarding domain registration, I own auroranrunner.com and manage its DNS settings via the AWS console. Given the dynamic nature of my IP address, I employ the dy.fi service to update the DNS record for my dy.fi domain automatically. On AWS Route 53, ha.auroranrunner.com is configured with a CNAME record pointing to sirius.dy.fi, a nifty setup. Thanks to my router’s dy.fi support, any alterations to my external IP are automatically synchronized.

Step 5: Troubleshooting Common SSL Setup Issues

While setting up SSL on HomeAssistant using Let’s Encrypt Certbot is straightforward, you might encounter some issues along the way. One common issue is the “Failed authorization procedure” error. This usually occurs when Certbot is unable to verify domain ownership. To resolve this, you need to ensure that your domain name is correctly pointed to your HomeAssistant’s IP address.

Another common issue is the “SSL connection error”. This usually occurs when HomeAssistant is not correctly configured to use the SSL certificate. To resolve this, you need to ensure that the paths to the SSL certificate and its corresponding private key in your HomeAssistant configuration file are correct.

Setting up SSL on HomeAssistant using Let’s Encrypt Certbot is a good way to secure your system. While the process might seem complex, it can be broken down into five easy steps: installing Certbot, generating an SSL certificate, setting up SSL on HomeAssistant, configuring HomeAssistant with the SSL certificate, and troubleshooting common SSL setup issues. By following these steps, you can secure your HomeAssistant and ensure that your data remains safe and private.

Conclusion

Implementing SSL with Certbot is relatively straightforward for those who are well-acquainted with their network setup. This approach offers a security advantage over depending on third-party VPN solutions, which merely introduce an additional layer to your existing infrastructure. Leveraging third-party services to manage your smart home system does not enhance security; rather, it compromises it. While VPNs can serve as a viable security measure for those lacking the expertise to properly configure their home networks, the assertion that third-party VPNs inherently bolster security is misleading.

For those considering a VPN, I advocate for hosting your own. In my experience, OpenVPN has been fully compatible with HomeAssistant, offering a cost-effective solution without the need for extra expenditures. Like the SSL setup, OpenVPN requires dynamic DNS unless you have the luxury of a static IP address, ensuring reliable and secure remote access to your smart home systems.

Home Assistant: Heat Pump Automation with Cheap SPOT hours and Github Copilot doing the work

Introduction

Finland has been part of Nord Pool, a pan-European power exchange, since 1998. Meaning, when you sign your power contract with electricity supplier, you can choose a contract utilising the power stock exchange prices.

The prices for the next day are announced every day around 1pm CET. You can combine this information for example with weather forecast to plan your electricity usage for the cheapest hours where applicable.

Home Assistant on the other hand has Nord Pool integration which enables you to optimise the electricity SPOT pricing. There is a lot of articles on how to do that to help you to get started. This articles goes through my current setup and my own experience with both Home Assistant and electricity stock pricing. And how I made everything working with GitHub Copilot vim plugin.

Typical claim is, that normal user cannot really utilise the power stock pricing since it is too much work, warming up the house takes constant amount of energy so there is no way to optimise or it is too much work to do the automation in he first place. The latter might be true, but if you take building a smart home as a hobby, then even that is not true. The more time it takes, the more fun it is.

Home Assistant is a hobby anyway. It’s non commercial product and it is Cloud independent: Meaning, you set it your yourself and you maintain it yourself in your own server. That being said, it is fairly easy to set up. You just need to have a server to install it. That can be dedicated server or mini computer like Raspberry Pi, old PC you have no other use or something that can run Linux.

My choice was to to use my Asus PN41 mini PC I already had running Ubuntu which I had set up earlier to run as my sandbox having several virtual machines running in it. Instead of adding another virtual machine I decided to setup Home Assistant as Docker Container. Installation and set up did not really take too long time. Once I installed mobile app to my phone I already had working setup.

The reason why I wanted to have Home Assistant in the first place though is, that I had two Toshiba Shorai Edge heat pump internal units installed, and Toshiba’s mobile app is installable only with European apple id. I have North American apple id and I really cannot change that, since although living most of the time in Europe, I have close ties to North America. After some googling I figured out that I can get around the limitation with this totally new thing for me at the time called Home Assistant.

Not only did I get the heat pump controls work with Toshiba AC integration I also got the Nord Pool spot prices available on nice ApexCharts and even predict for Nord Pool prices relying on Random forest machine learning algorithms as illustrated below.

After I had Home Assistant container running, Toshiba AC integration installed and mobile app on my phone, I was good to go. Setup up is really fast to do as long as one is familiar with the related technology it really doesn’t take more than an hour. My initial aim was just to be able to manage the internal heating units through my phone. Then later I noticed that ok, it is also much easier, for example, to schedule the heat pumps to different temperatures different times with Home Assistant than with extremely cumbersome Toshiba remote.

On the other hand, I noticed Home Assistant itself had plenty of other interesting features I could utilise while building a smart home gradually. I got four Shelly H&T and one  Shelly Plus H&T thermometers I could have on my Home Assistant dashboard. Three Shelly Plugs to monitor electricity usage for the Heat Pump and other appliances.

Automation

Just having Home Assistant Mobile App running enabled me being able to control heat pump units, follow room temperatures, current weather and forecast, electricity consumption and price is of course nice, but everything is still done manually. I felt I’m missing at least half of the benefits and nothing really changed anything yet.

Then I found this blog post on how to automate device for cheapest hours and it was pretty much all I was looking for. At least on idea level it was. It grabs the next days cheapest electricity prices and one can schedule heat pump to increase temperature when the electricity is on it’s cheapest. This happens typically at night – it is just after midnight almost always. I wasn’t very familiar with yaml and I still find the syntax cumbersome to get anything working – anything working easily at least. There’s plenty of scheduling solution with GUI based forms, but for me understanding those was even more difficult. I got this solution for getting next day’s cheapest hours and increase heating during them to work fine except for one thing. Once it started, it did not stop without manually stopping it.

I decided to create a schedule which set the heating back from 24C to 20C at 5am. With Home Assistant of course. If the cheapest hours are at day time, that does not work though. But it worked well enough almost for a year. Then I got more involved with yaml while learning Ansible and writing Pipelines for Azure with yaml. I also utilised yaml syntax highlighting on vim, so it all started to get easier.

Why write own code when there’s Github Copilot

Completing the first idea

The biggest motivator I found was Github Copilot. I started to use it while writing Python code, but noticed it helps quite a lot with yaml too. I only wanted to change my automations.yaml slightly. I wanted to get the part working, where the heating should stop. And I don’t want any heating blowing full 24C during day time either. Copilot does not write it to you, but it makes it easier to get it done.

So I did this: added the time conditions with after and before.


# Set temp to 24C when the SPOT price is at it's cheapest.
- id: '1663399614818'
  alias: Increase heating
  description: 'Cheap energy time set heating to 24C'
  trigger:
  - platform: time
    at: input_datetime.device_start_time
  condition: 
    condition: and
    conditions:
      - condition: time
        after: '00:00'
        before: '04:00'
  action:
  - service: climate.set_temperature
    data:
      temperature: 24
    target:
      entity_id: climate.ac_12494102
  mode: single

The code without timing conditions are available from the blog post link above, so I’m not writing it here, although you can check my full automations.yaml from my GitHub repo – not that I expect it to help anyone or to be perfect, but there it is. Then next thing is to stop the increased heating. To be noted, I constantly work on my automations, so the code in repo does not necessarily reflect what I have demonstrated here.

# Set temp to 20C at end of cheap hours
- id: '1663399614821'
  alias: Hallway AC temp to 20
  description: 'Cheap energy end time set temp to 20'
  trigger:
  - platform: time
    at: input_datetime.device_end_time
  condition: 
    condition: and
    conditions:
      - condition: time
        after: '03:00'
        before: '06:00'
  action:
  - service: climate.set_temperature
    data:
      temperature: 20
    target:
      entity_id: climate.ac_12494102
  mode: single

I didn’t have time conditions there as time of writing this, but I added them later once I had verified everything works correctly. With Home Assistant it’s better to build things gradually. Then you know easier what does not work and what does.

I also wanted to have things like: If electricity is more expensive than 15c/kWh, decrease heating by 1C:

# If SPOT price is above average let's set heating 1C lower
- id: hallway_ac_fan_expensive_spot
  alias: If spot price above average cents set heat 1C lower
  description: ''
  trigger:
  - platform: numeric_state
    entity_id: sensor.nordpool_kwh_fi_eur_3_10_024
    above: sensor.energy_spot_average_price
  condition: 
    condition: and
    conditions:
      - condition: time
        after: '08:00'
        before: '22:00'
  action:
  - service: climate.set_temperature
    data:
      temperature: "{{ state_attr('climate.ac_12494102', 'temperature') - 1 }}"  # Decrease temperature by 1 degree
    target:
      entity_id: climate.ac_12494102
  mode: single

The above is partly written by ChatGPT, but it typically generates code, which needs a lot of tweaking to get it to work for real, but some of it is usable.

I also often turn heater off when outside is a bit warmer and don’t necessarily remember to put it on before going to sleep. At least in theory this could lead to situation where it gets really cold at night, and then the heater is off when temperature is way below 0C. Then one should really not turn it on anymore before it gets warmer, since it decreases the life of the outside unit some what. If not significantly even.

# If outdoor temp is below 1C turn on hallway AC
- id: hallway_ac_fan_on_low_temp
  alias: If temp below 1 set on
  description: ''
  trigger:
  - platform: numeric_state
    entity_id: sensor.ac_12488762_outdoor_temperature
    below: 1
  condition: []
  action:
  - service: climate.turn_on
    target:
      entity_id: climate.ac_12494102
  mode: single

Expanding the ideas

Above was just first step though. I wanted to have more. Simple things though. I struggled a day with getting my next idea to work. The idea is simple:

  • Increase heat, when spot price is above daily average.
  • Decrease heat, when spot price is below daily average.

I had everything working with fixed values. But daily average spot price varies a lot, so I’m not ok with fixed value. I tried to use something like state_attr('sensor.nordpool_kwh_fi_eur_3_10_024', 'average'). Looks valid to me, but when I tried to use it, it just didn’t work. I tried to “cast” since I always got error “could not convert string to float” no matter what I trid.

Then I figured out just by myself with no Github Copilot, that if I put above to sensors.yaml and create a sensor having the daily average, I might be able to use that. Bingo!

energy_spot_average_price:
      friendly_name: "Nordpool Average Spot Price"
      unit_of_measurement: 'c/kWh'
      value_template: "{{ state_attr('sensor.nordpool_kwh_fi_eur_3_10_024', 'average') | float | round(2) }}"

Above I have created sensor: sensor.energy_spot_average_price on sensors.yaml. That I can use on automations.yaml as shown below:

# If outdoor temp is below 1C turn on hallway AC
- id: hallway_ac_fan_on_low_temp
  alias: If temp below 1 set on
  description: ''
  trigger:
  - platform: numeric_state
    entity_id: sensor.ac_12488762_outdoor_temperature
    below: 1
  condition: []
  action:
  - service: climate.turn_on
    target:
      entity_id: climate.ac_12494102
  mode: single

Since I’m increasing heat above I want to do it only when it’s relatively cold outside. Also I want to do it only during day, when the prior cheapest prices logic is not active. That is why I have set this to do following:

  • Between 10am and 5pm:
    • When outside temperature is below +2C and spot price is below daily average:
      • Lower the heat on Hallway AC by 1 degree celcius

Then I have another entry for decreasing the heat, when spot price goes above daily average. For that I don’t use the requirement for outside temperature, since if it’s warmer than that, I’m always ok to decrease the temperature.

Purpose

The goal for me is to heat a bit more when electricity is cheaper and then heat a bit less when it’s more expensive. Air is not very good on preserving the heat, but it does it a bit. Also, when I go to sleep, I don’t need to heat. My house colder at least till midnight since there’s almost no heating. The after midnight there’s typically the cheapest hours in hand and my system starts to overheat a bit. Pretty normal pattern is, that when I wake up, the electricity price starts to go up during the normal morning hours when other people wake up as well. My heating system isn’t really needed by then and the temperature starts going down gradually till it is needed gain.

Rest of the day my system follows the strategy to lower heat slightly if price goes above average and heat a bit more when it the price goes below average

This will optimize the heating the way, that most of the time the average price I pay for electricity is bit lower than the average spot price, which is my intention.

Below pictures shows, how the heating takes in place at midnight. The stops at 4am. Next hike is around 6am, when the upstairs heat pump in bedroom is turned on after waking up. The bedroom heating is never on during night and most of the automation is only for Hallway AC.

The yaml code needed

The examples here are pretty much copy/pasted from Toni’s blog post so credits to him.

configuration.yaml

Home Assistant needs a configuration file configuration.yaml and there you need following to get the cheapest hours utilized.

# Helper to keep the start time
input_datetime:
  device_start_time:
    name: Device Start Time
    has_time: true
    has_date: false
  device_end_time:
    name: Device End Time
    has_time: true
    has_date: false
# Include automations.yaml and sensors.yaml
automation: !include automations.yaml
sensor: !include sensors.yaml                                                                                                                                          

sensors.yaml

On sensors.yaml you need following. Note that sensor.nordpool_kwh_fi_eur_3_10_024 must be replaced with the sensor you have for Nord Pool integration.

- platform: template                                                                                                                                                   
  sensors:
    energy_spot_average_price:
      friendly_name: "Nordpool Average Spot Price"
      unit_of_measurement: 'c/kWh'
      value_template: "{{ state_attr('sensor.nordpool_kwh_fi_eur_3_10_024', 'average') | float | round(2) }}"
 
    cheapest_hours_energy_tomorrow:
      device_class: timestamp
      friendly_name: Cheapest sequential electricity hours
      value_template: >
        {%- set numberOfSequentialHours = 3 -%} 
        {%- set lastHour = 23 -%} 
        {%- set firstHour = 0 -%} 
 
        {%- if state_attr('sensor.nordpool_kwh_fi_eur_3_10_024', 'tomorrow_valid') == true -%} 
          {%- set ns = namespace(counter=0, list=[], cheapestHour=today_at("00:00") + timedelta( hours = (24)), cheapestPrice=999.00) -%} 
          {%- for i in range(firstHour + numberOfSequentialHours, lastHour+1) -%} 
            {%- set ns.counter = 0.0 -%} 
            {%- for j in range(i-numberOfSequentialHours, i) -%} 
              {%- set ns.counter = ns.counter + state_attr('sensor.nordpool_kwh_fi_eur_3_10_024', 'tomorrow')[j] -%} 
            {%- endfor -%} 
            {%- set ns.list = ns.list + [ns.counter] -%} 
            {%- if ns.counter < ns.cheapestPrice -%} 
              {%- set ns.cheapestPrice = ns.counter -%} 
              {%- set ns.cheapestHour = today_at("00:00") + timedelta( hours = (24 + i - numberOfSequentialHours)) -%} 
            {%- endif -%} 
          {%- endfor -%} 
          {{ ns.cheapestHour }}
          {%- set ns.cheapestPrice = ns.cheapestPrice / numberOfSequentialHours -%} 
        {%- endif -%}                   

automations.yaml

Now Here are the triggers I have created in automations.yaml. I have three triggers for pumping up the heat with each one different action for cheap hours. Combining actions with one trigger seem not to work, or I don’t know correct syntax. I decrease the heat after four hours, but since I don’t need to stop heater, when the heating gets decreased. I have only two actions.

First I need to create the input_date times to use later:

# Set device start time: Needs cheapest_hours_energy_tomorrow in sensor.yaml                                                                                           
- id: '1663398489357'
  alias: 'Set device start time'
  description: ''
  trigger:
  - platform: time
    at: '23:10:00'
  condition:
  - condition: not 
    conditions:
    - condition: state
      entity_id: sensor.cheapest_hours_energy_tomorrow
      state: unknown
  action:
  - service: input_datetime.set_datetime
    data:
      time: '{{ as_timestamp(states(''sensor.cheapest_hours_energy_tomorrow'')) | timestamp_custom(''%H:%M'') }}'
    target:
      entity_id: input_datetime.device_start_time
 
 
# Set device end time 4 hours after start time: Needs cheapest_hours_energy_tomorrow in sensor.yaml
- id: '1663398489358'
  alias: 'Set device end time'
  description: ''
  trigger:
  - platform: time
    at: '23:15:00'
  condition:
  - condition: not 
    conditions:
    - condition: state
      entity_id: sensor.cheapest_hours_energy_tomorrow
      state: unknown
  action:
  - service: input_datetime.set_datetime
    data:
      time: '{{ ((as_timestamp(states(''sensor.cheapest_hours_energy_tomorrow'')) + (3600*4)) | timestamp_custom(''%H:%M'')) }}'
    target:
      entity_id: input_datetime.device_end_time
  mode: single

Then the actual triggers:

# Do the actions when time trigger is hit.
# Each action separately: Turn on, set temp, set fan mode
# Make sure AC is on before setting temp or fan mode
- id: '1663399614817'
  alias: Turn on Hallway AC
  description: 'Cheap energy time turn on hallway AC'
  trigger:
  - platform: time
    at: input_datetime.device_start_time
  condition: 
    condition: and
    conditions:
      - condition: time
        after: '00:00'
        before: '05:00'
  action:
  - service: climate.turn_on
    target:
      entity_id: climate.ac_12494102
  mode: single
  
# Set temp to 24C
- id: '1663399614818'
  alias: Increase heating
  description: 'Cheap energy time set heating to 24C'
  trigger:
  - platform: time
    at: input_datetime.device_start_time
  condition: 
    condition: and
    conditions:
      - condition: time
        after: '00:00'
        before: '05:00'
  action:
  - service: climate.set_temperature
    data:
      temperature: 24
    target:
      entity_id: climate.ac_12494102
  mode: single
# Set fan mode to high    
- id: '1663399614819'
  alias: Hallway AC fan to high
  description: 'Cheap energy time set fan to high'
  trigger:
  - platform: time
    at: input_datetime.device_start_time
  condition: 
    condition: and
    conditions:
      - condition: time
        after: '00:00'
        before: '05:00'
  action:
  - service: climate.set_fan_mode
    data:
      fan_mode: "High"
    target:
      entity_id: climate.ac_12494102
  mode: single
  
# Lower fan from High to Auto four hours after start time
- id: '1663399614820'
  alias: Hallway AC fan to Auto
  description: 'Cheap energy time set fan to Auto'
  trigger:
  - platform: time
    at: input_datetime.device_end_time
  condition: 
    condition: and
    conditions:
      - condition: time
        after: '04:00'
        before: '09:00'
  action:
  - service: climate.set_fan_mode
    data:
      fan_mode: "Auto"
    target:
      entity_id: climate.ac_12494102
  mode: single
# Set temp to 20C four hours after start time
- id: '1663399614821'
  alias: Hallway AC temp to 20
  description: 'Cheap energy time set temp to 20'
  trigger:
  - platform: time
    at: input_datetime.device_end_time
  condition: 
    condition: and
    conditions:
      - condition: time
        after: '04:00'
        before: '09:00'
  action:
  - service: climate.set_temperature
    data:
      temperature: 20
    target:
      entity_id: climate.ac_12494102
  mode: single
# If SPOT price is below average we can increase heating by 1C during day time
- id: hallway_ac_fan_low_spot
  alias: If spot price below 7 cents increase heat
  description: 'With low price increase heat by 1'
  trigger:
  - platform: numeric_state
    entity_id: sensor.nordpool_kwh_fi_eur_3_10_024
    below: sensor.energy_spot_average_price
  - platform: numeric_state
    entity_id: sensor.ac_12488762_outdoor_temperature
    below: 2 
  condition: 
    condition: and
    conditions:
      - condition: time
        after: '10:00'
        before: '17:00'
  action:
  - service: climate.set_temperature
    data:
      temperature: "{{ state_attr('climate.ac_12494102', 'temperature') + 1 }}"  # Increase temperature by 1 degree
    target:
      entity_id: climate.ac_12494102
  mode: single
# If SPOT price is above average let's set heating 1C lower.
- id: hallway_ac_fan_expensive_spot
  alias: If spot price above 7 cents set heat 1C lower
  description: ''
  trigger:
  - platform: numeric_state
    entity_id: sensor.nordpool_kwh_fi_eur_3_10_024
    above: sensor.energy_spot_average_price
  condition: 
    condition: and
    conditions:
      - condition: time
        after: '10:00'
        before: '17:00'
  action:
  - service: climate.set_temperature
    data:
      temperature: "{{ state_attr('climate.ac_12494102', 'temperature') - 1 }}"  # Decrease temperature by 1 degree
    target:
      entity_id: climate.ac_12494102
  mode: single

Full examples

My full yaml files are also in my personal GitHub repo:

Summary

Home Assistant is useful tool to make some simple home automations. Obviously getting the heat pump itself have saved me plenty on electricity bills, but Home Assistant takes me one step further.

Although Home Assistant does provide nice GUI for creating schedules, I do prefer editing the text based yaml files. yaml itself is error prone format and for that good editor is a must. My choice of editor has been vim for last 20 years at least and I see no reason to switch away from it. Although I have tried to switch to Eclipse, Pycharm, VS Code – yet I always go back to vim. I even tried neovim but couldn’t find any difference compared to vim (I do not use lua).

When I found Github Pilot plugin for vim I found it to be a game changer. Not only for writing Python and Azure Pipelines with yaml, but especially for Home Assistant configuration yaml files. I also feel GitHub Copilot extremely addictive. The way it provides suggestions makes me chuckle once in a while and I really miss it almost everywhere – almost. It really would need to write my commit messages with vim fugitive. Feature suggestion for Tim.

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.

Db2

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

Oracle

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

MySQL: TEST RESULT: 1739 NOPM from 5252 MySQL 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.

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: https://www.virtualbox.org/wiki/Downloads
  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: https://www.centos.org/download/
  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 192.168.56.101. 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/VBoxLinuxAdditions.run
  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 https://centos7.iuscommunity.org/ius-release.rpm
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
./unpack
cd ../linux64
./unpack

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):

NZ_HOST=netezza.domain.com
NZ_DATABASE=SYSTEM
NZ_USER=admin
NZ_PASSWORD=password
export NZ_HOST NZ_DATABASE NZ_USER NZ_PASSWORD
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/nz/lib64
export PATH=$PATH:/usr/local/nz/bin
export ODBCINI=$HOME/.odbc.ini
export NZ_ODBC_INI_PATH=$HOME

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
SYSTEM.ADMIN(ADMIN)=>

Setup ODBC

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

odbc.ini
odbcinst.ini

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

 Questions

If you have any questions, please connect with me.

Update

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.