whatsuphome

I stumbled upon something called Nanobrowser. It's a Chrome extension which turns your browser into open source AI web agent. 

What's an AI web agent? It makes your web browsing more exciting, with all the gotchas of LLMs combined with the speed and reliability that resembles a drunk grandpa attempting to use a computer without his glasses. No, jokes aside, it is kind of impressive. You just tell it what you want and it attempts to use your browser instead of you. Check this out:

I admit, this entry is very far-fetched (or near-fetched, depending on your camera position), but I'm sure you can come up with actual use cases for this. One scenario I can think of is that if you have a CCTV in your factory or whatever and something should stay within X meters from the camera, you can do it with this trick.

Zabbix 7.4.0rc1 came out and of course I upgraded my What's up, home? Zabbix version to it. Even though the changes in Zabbix 7.4 seem to be relatively minor, there's some very welcome changes and new features. Underneath, there's good stuff for bigger environments: ability to give bigger history cache, trend cache; database query optimizations; your usual set of bugfixes and performance improvements ...  but there's more than that.

Lately my home router ASUS AX-68U with Asuswrt-Merlin firmware has been using much more CPU than it used to do. It's not being slow or anything, but I'm curious how its CPU usage jumped since I last updated the firmware. This kind of jump in CPU usage cannot be a coincidence and definitely something that I want to track down. Below is a graph from my home Zabbix dashboard.

First, apologies for not posting in a while, I have been terribly busy at work and at home as we have the puppy.

When playing around with different LLMs with Ollama, you might want to see how fast some model is. Ollama won't return you any performance statistics in server mode (or does it? Let me know!), so to get the statistics to Zabbix I had to be creative. For those who don't know, Ollama is a cross-platform software for running all kinds of LLMs locally.

OpenAI's latest ChatGPT 4o image generator has been all the rave lately. But did you know it can help you to generate Zabbix templates and other very technical stuff in the silliest way possible - by just drafting something on paper? This is not perfect yet, coming with all the gotchas that you have with the code generated by LLMs, but oh boy how we live in the future. Observe!

The original drawing

This was me last night just typing... no what's the old-fashioned way... writing on a paper. I feel bad for you as you try to decipher my handwriting. 

Earlier this year, Maikki the poodle joined our family, and it's about time that we start monitoring her. This time the idea came from my wife, as she spotted Invoxia Minitailz smart AI collar for dogs during some dog training session she attended to. Yesterday Maikki entered the sweet, sweet world of monitoring.

Many years ago, maybe around 2017 or 2018, one of my ex-colleagues (Hi, Kevin!) said that I probably would use Zabbix even to come up with the lottery numbers. Back then, just to strike back, I did exactly that, with a small easter egg in work Zabbix containing the lotto numbers. That was a quick bash script feeding the Zabbix item.

Now, let's return to that topic, but use Zabbix Script item type instead. Also, let's take a look at few other details that help in monitoring.

Wouldn't it be fancy if your Zabbix could tell you if your Raspberry Pi would have a pending EEPROM update waiting for you? It's so easy just to upgrade all the packages and forget to update the EEPROM with sudo rpi-eeprom-update -a. 

Luckily there's a simple way to make Zabbix to check this for you.

Would you like to get AI added to your Zabbix now? Like, RIGHT NOW? In few minutes with no account, no payment, no server, no scripting, no strings attached? 

One of the monZphere's modules that easily flies under the radar without nobody noticing would be their AIModuleMonzphere, freely available from GitHub. 

What makes it different? Most likely your AI additions call home to their mothership, or at the very least consult their local LLM server hosted somewhere in your network.