On the main interface, it does not filter out the tagged networks packets. It sends tagged and untagged packets to the VM’s that are on that interface.
When you need a VM to communicate on different vlans, it is easy for it to work.
When you want your VM to be isolated, you need to have it on a vlan. This is a Very important thing to be aware of.
For a better security posture, no VM should be on the main interface in XCPNG.
I have NetAlertX up and running. It is configured to see all devices on all networks (vlans) from ARP requests.
Still to do.
- Configure devices and tweak settings
- Set up notifications for external alerting
To better understand AI more, I am wanting to host a dedicated AI vm, and learn how to use it through out my home lab. One of the ideas I have it to have it be able to speak and listen.
Other ideas are using it to sort my picture albums, watch system logs, verify scripts…
One project that I have been pondering for a while is using Home Assistant in my home/home lab to have some home automatons (lights, water plants, AC, Thermostat) and integrate an LLM for voice and audio interactions. In a nut shell, make my own “alex, siri” and keep it all isolated to my home and on my network.
This will be a fun and informative experience. It is also a lot of small pieces, it can be built out slowly, bit by bit over time.
Currently on my laptop and on my desktop, I am running Ollama, with GPU acceleration, and Page Assist (web browser plug in) to help me with my studying and finding of information.
I am finding it helpful to ask it a question, such as, “I know how to install LDAP, can you tell me where I can find HOW to use LDAP. I have no experience, apart from installing it.” and it will search the web and find resources to help me learn. I can then ask it more questions, and it will refine its response from there.
I am starting to use it more, so I can keep my mind focused on the topic I am trying to learn/figure out, and not get distracted/unmotivated/frustrated with being unable to find that one resource that I am in need of.
I have also used it to help me understand topics, like Regular Expressions. I ask it to provide an example of one and then explain it. This is one way that I can gain a deeper understanding of the topic. It is learning how it works, seeing it in examples, then trying it on my own with out the pressure of a “homework” timeline or a teacher that has no time or patients.
AI will always be an assistant, something to render help. (Just like the hardware that it has run on for decades.) We must be be diligent with its output and always exercise discernment with it, just like with any person that we interact with in our daily lives.
I have an alert and notification in Graylog that sends me a slack message when one of my systems runs Alsible pull. (Pulled from Sys Logs)
I do not need a fancy or in depth message, just that X host has run. Here is my custom message.
${foreach event.fields field}
${field.key}: ${field.value} ran ansible at ${event.timestamp}
${end}
I also have a custom field set up under the alert as
template: “${source.source}”require_values: true
When I have developed my Ansible playbooks more, I might implement this into my ansible deployment.
https://xen-orchestra.com/blog/virtops3-ansible-with-xen-orchestra
The services that I am planning/pondering self hosting in my home lab.
Core Services
WAN Services
LAN Services
- Podman, or something along the same lines.
- Ollama with webUI (On a dedicated VM with GPU acceleration)
- An eBook reader.
This is a current list of the services that I have running in my home lab.
Core Services
- GrayLog
- HAProxy
- Omada
- Xen Orchestra
- SyncThing
WAN Services
LAN Services
- GitLab
- FreshRSS
- NetAlert (in progress)
- A Ticketing service (in progress, in testing)