https://lonesysadmin.net/2013/12/22/better-linux-disk-caching-performance-vm-dirty_ratio/
https://github.com/intel/compute-runtime
https://tomthegreat.com/blog/setting-up-ubuntu-20-04-lts-for-plex-with-intel-gen-12-cpu/amp/
Tailscale races DNS so by using their DNS implementation and configuring multiple DNS servers you can easily create a highly available DNS setup.
- Install Adguard on 2 (or more) hosts and configure
- Install Adguardhome-sync and configure
- Install Tailscale on each host and configure, make sure to accept dns (
tailscale up --accept-dns) - Add each Adgaurd host Tailscale IP to Tailscale DNS (through the dashboard) and ensure you toggle "Override local DNS" to On
Tailscale will now query all of the DNS servers and return whichever result comes back fastest. Adguard sync means you can set a primary server and have everything automatically updated.
One thing I choose to do is not sync DNS rewrites and create a domain called test.test and enter the Tailscale URL for that host so I can easily query test.test and see which host is responding to my request.
Modern releases of Ubuntu (17.10+) and Fedora (33+) include systemd-resolved which is configured by default to implement a caching DNS stub resolver. This will prevent pi-hole from listening on port 53.
sudo sed -r -i.orig 's/#?DNSStubListener=yes/DNSStubListener=no/g' /etc/systemd/resolved.conf
sudo sh -c 'rm /etc/resolv.conf && ln -s /run/systemd/resolve/resolv.conf /etc/resolv.conf'
sudo systemctl restart systemd-resolved
Credit to pihole for this: https://github.com/pi-hole/docker-pi-hole/?tab=readme-ov-file#installing-on-ubuntu-or-fedora
Monitor and respond to failures automatically: https://github.com/hucknz/useful/blob/main/monit.md
SnapRAID helper script to run each day and notify of outcomes via healthchecks.io: https://github.com/hucknz/useful/tree/main/snapraid
Some Linux operating systems have moved to systemd-journald for logging. Out of the box New Relic doesn't seem to capture these logs and fluent-bit doesn't support a wildcard. To add all the systemd units to New Relic you can run the following script to generate the config and add it to your logging.yml file.
cd scripts
wget https://raw.githubusercontent.com/hucknz/useful/refs/heads/main/systemd_log_generator.sh
sudo chmod +x systemd_log_generator.sh
sudo ./systemd_log_generator.sh
sudo nano /etc/newrelic-infra/logging.d/logging.yml
Credit to mlemieux and szilvesztercsab+nrtest0 in this thread: https://support.newrelic.com/s/hubtopic/aAX8W000000L5ghWAC/systemd-logging-how-to-do-all