Server

I have recently seen lost of AI content from many big companies and have wanted to experiment with it my self. The most practical thing I have been able to think of is an AI assistant on my PC. I started when I saw this video (https://www.youtube.com/watch?v=XvbVePuP7NY&t) from network chuck. This was prity close to what I wanted so I gave it a go. The first issue I ran into was the inability to have add ons when home assistant is installed in docker. However this was a simple fix I just installed them as separate docker containers and manually added them to the whyoming intergration. I then installed the whyoming satalite on my main PC which went smoothly. However I did the extra step of adding sounds to the start and end. ...

It started when I went back to my pterodactyl panel after a long time. I saw a red heart and a invalid ssl cert warning in console. I renewed the cert and the heart went green. However my servers still wern’t loading. So I changed the forwarding from cloudflare to nginx proxy. This still didn’t work. Then I renewed the cert as rsa type and noticed it put it a folder that ended in 00001. So I deleted the one with out the numbers and then renamed the one left to remove the numbers. One restart later and it was all working.

During a backup test I did recently I found that wireguard wasn’t working properly. I have no idea how long this has been the case but to fix it I started trying to change the config however when that lead to no improvment I decided to reinstall. This is something I have considered before as my current setup doesn’t have a gui which would be nice so I use the repository (https://github.com/wg-easy/wg-easy?tab=readme-ov-file). The install was over all easy however the password hashing didn’t work on a super strong password. This seems to be working now and I will probably redo the backup test for wireguard soon.

When I revisted my kasm server I found I was unable to launch a session. After running docker ps I saw that the guac container was contantly restarting. After searching the error message I found this link (https://kasmweb.atlassian.net/servicedesk/customer/portal/3/article/8126468) which shows a fix. After following this guide I got it working again. However the same day one of my friends tried to download a game on a steam continer and I don’t know why but the entire vm froze up and I had to hard stop it. When I restarted it I found the config files filled with @ and nothing else. ...

I noticed today that Wazuh wasn’t working after I restarted my server. I saw in the logs it was failing the bootstrap test. I thought the issue was the mass of file permission error that were appearing before this error showed. However after messing around with file permissions I noticed a error about sysctl vm.max_map_count after running “sysctl -w vm.max_map_count=262144” it started working however I have no idea how this wasn’t a issue until now.

I decided to finally remove that annoying pop up that appears every time I log into my server. I found this guide (https://www.youtube.com/watch?v=5ftWGupG66A). i ignored the first section about the repositories but the section about removing the warning was clear and easy to follow. Once I had restarted the service I went to check the web UI and I was still getting the warning so I checked in a incognito tab and saw it was gone I just need to wait for my cache to update.

I was originaly planing to try security onion but while reasurching it I came across Wazuh which runs on docker which I always prefer. Once I installed it using a edited version of the compose file in the github repository. I managed to install the agent on two of my servers however on my main proxmox host I didn’t understand that the server I was the ip of the main host not the server I was installing it on. This lead to a botched install I haven’t been able to fix yet. ...

One day when I tried to use the console section of portainer I got the error “Unable to retrieve image details”. After looking online I saw that I needed to update portainer so I started looking for the compose file and couldn’t find it. Well that was a bit annoying so I ran “docker run –rm -v /var/run/docker.sock:/var/run/docker.sock ghcr.io/red5d/docker-autocompose …” which worked as expected. I then pulled the latest image and trimed the compose file. When I removed and restarted the container I was annoyed to see it still wasn’t working. ...

This has for a long time been probably my second biggest issue with my homelab. My main website that your reading this on has been running on http. This happened because i use the domain keldo.uk which is also the domain I use for email. This ment I couldn’t route through cloudflare like everything else. My original attempt to fix this was with trafik but when I got as far as forwarding the website all I got was a white page. I presumed that nginx proxy manager would be the same however I was proven wrong when I tried today. ...

I started with the docker compose file from https://github.com/open-webui/open-webui. I then edited this changing the volumes to mounts on my nas and changing the port. I then installed a range of models some designed for general conversations and others for coding. I then wanted to add tab auto complete to vscode so i used a section of technotim’s guide (https://technotim.live/posts/ai-stack-tutorial/). This worked fine and was all round a simple process. I look forward to trying to customise some of the models.