.noise

From Noisebridge
Revision as of 11:53, 1 July 2025 by 2a10:8a40:f313::29 (talk) (Enhanced writeup)
Jump to navigation Jump to search

.noise - Noisebridge's Custom TLD

This TLD works on Noisebridge's WiFi network(s), providing convenient DNS names for hosts within the subnet 10.21.0.0/23 (10.21.0.0 - 10.21.1.255), which we use for both the WiFi network and internal LAN. For more details, see Resources/Network.

Several servers are hosted at noisebridge with the `.noise` TLD:

    1. Servers
  • einhorn.noise hosts some VMs (unicorn in German; not to be confused with the former server Unicorn).
  • ml1.noise and ml2.noise provide access to shared-use GPUs.
  • Olympus.noise manages the door access control system.

To verify that your host can resolve the .noise TLD to the local network's DNS server, you can perform the following tests:

```

  1. !/bin/bash
  1. Check if .noise TLD resolves with default DNS

if nslookup example.noise > /dev/null; then

 echo ".noise TLD resolves with default DNS"

else

 # If not, test if querying 10.21.0.1 will resolve it
 if nslookup example.noise 10.21.0.1 > /dev/null; then
   echo ".noise TLD resolves when querying 10.21.0.1"
 else
   echo ".noise TLD does not resolve with default DNS or 10.21.0.1"
 fi

fi ```

For hosts with hardened DNS settings, DNS query exemptions can be configured using dnsmasq to route requests for .noise domain resolution exclusively to the gateway router's DNS service at 10.21.0.1. Here is an example configuration: ```

  1. Configure dnsmasq to use the local DNS server for .noise domains

server=/noise/10.21.0.1

  1. Optional: specify the IP address of the dnsmasq server

listen-address=10.21.0.100 ``` To apply this configuration, follow these steps:

1. Install dnsmasq on your system 2. Create a configuration file (e.g., /etc/dnsmasq.conf) and add the above configuration lines. 3. Restart the dnsmasq service to apply the changes.

With this configuration in place, your host will use the local DNS server for .noise domains, providing an additional layer of security. Contributions with further configuration examples, including GPT-generated output, are welcome.