Skip to content

Instantly share code, notes, and snippets.

@besmirzanaj
Last active March 13, 2025 09:40
Show Gist options
  • Save besmirzanaj/490c7f8ff61f6e9681fa5656220e3910 to your computer and use it in GitHub Desktop.
Save besmirzanaj/490c7f8ff61f6e9681fa5656220e3910 to your computer and use it in GitHub Desktop.
synch two technitium servers
#!/bin/bash
# Author: Besmir Zanaj, 2024
# This is a very raw script to backup configs (no logs and no stats) from a technitium server
# to another
#
# first create two tokens: one on the source server and another one on the destination one
# fill out the vars below
# create a cronjob with this script on the destinaton host
# eg:
# 30 */6 * * * /path-to/technitium-sync.sh
set -euxo pipefail
src_dns_server='source.ip.address'
dst_dns_server='dest.ip.address'
src_dns_serverdomain='fqdn.of.source.server'
dst_dns_serverdomain='fqdn.of.dest.server'
src_dns_token='SOURCE_TECHNITIUM_TOKEN_HERE'
dst_dns_token='DEST_TECHNITIUM_TOKEN_HERE'
backup_file="/tmp/technitium-backup.zip"
# update the dhcp scope as per your local settings
dhcp_scope_name="local-home"
# Ensure required tools are installed
command -v curl >/dev/null 2>&1 || { echo "curl is not installed. Aborting." >&2; exit 1; }
# Check the primary server's health before running the script
echo "Checking primary Technitium server status"
status_code=$(curl --write-out %{http_code} --silent --output /dev/null http://$src_dns_server:5380)
if [[ "$status_code" -ne 200 ]] ; then
echo "Primary DNS server is not available. Skipping backup"
exit 1
else
echo "Getting the backup archive from the primary server"
curl -s "http://$src_dns_server:5380/api/settings/backup?token=$src_dns_token&blockLists=true&logs=false&scopes=true&stats=false&zones=true&allowedZones=true&blockedZones=true&dnsSettings=true&logSettings=true&authConfig=true&apps=true" -o $backup_file
fi
# restore_backup
if [[ -f "$backup_file" ]]; then
echo "Restoring the backup on $HOSTNAME"
curl -s --form file="@$backup_file" "http://$dst_dns_server:5380/api/settings/restore?token=$dst_dns_token&blockLists=true&logs=true&scopes=true&stats=true&apps=true&zones=true&allowedZones=true&blockedZones=true&dnsSettings=true&logSettings=true&deleteExistingFiles=true&authConfig=true" --output /dev/null
# wait for server to come back
echo "Waiting for 10 seconds for the destination server to start up"
sleep 10
# set dnsServerDomain on destination server
echo "Updating DNS server Domain in destination server"
curl -X POST "http://$dst_dns_server:5380/api/settings/set?token=$dst_dns_token&dnsServerDomain=$dst_dns_serverdomain"
# disable DHCP on the destination server
echo "disabling DHCP in destination server"
curl -X POST "http://$dst_dns_server:5380/api/dhcp/scopes/disable?token=$dst_dns_token&name=$dhcp_scope_name"
# cleanup
echo "Cleaning up temporary files"
rm -rf $backup_file
fi
@bdorr1105
Copy link

Thanks for putting this thing together, I was sad to see this capability did not exist in technetium. Anyways, I think you could approach it a bit differently. If you put the NS and A record of the the secondary DNS server you can get it to sync the records without the script. That makes this a whole lot less complex. I went and looked at the available API options and focused it enabling and disabling DHCP based on whether the primary server is up.

I modified what you already put together and added some labeling in the script. I also opted to have it run a check every 5 minutes, but may bump that to an hour or so. This makes it pretty HA to me as the secondary automatically sync outside the script. I also went with the docker version as that is way more resilient to Operating System patches and updates and it can be ported to any VM or server

#!/bin/bash

# Author: Besmir Zanaj, 2024
# Modifed By: Brian Dorr
# This script checks the status of the primary Technitium server and enables or disables DHCP 
# on the secondary server based on the primary server's availability.
#
# Fill out the vars below.
# Create a cronjob with this script on the secondary host.
# eg:
# */5 * * * * /root/technitium-dhcp-failover.sh

set -euo pipefail

src_dns_server='primary.dnsserver.ip.address' # Primary server's IP address
dst_dns_server='secondary.dnsserver.ip.address' # Secondary server's IP address
src_dns_token='primay-dns-server-API-token' # Primary server's API
dst_dns_token='primay-dns-server-API-token' # Secondary server's API

# DHCP scopes to manage - puth the name of each scope you have
dhcp_scopes=("scope1" "scope2")  #Use this array for one or many scopes

echo "Checking primary Technitium server status"
status_code=$(curl --write-out '%{http_code}' --silent --output /dev/null http://$src_dns_server:5380)

if [[ "$status_code" -ne 200 ]]; then
  echo "Primary DNS/DHCP server is not available. Enabling DHCP on the secondary server."
  action="enable"
else
  echo "Primary DNS/DHCP server is available. Disabling DHCP on the secondary server."
  action="disable"
fi

for scope in "${dhcp_scopes[@]}"; do
  echo "Executing API call to $action DHCP scope: $scope"
  response=$(curl -X POST "http://$dst_dns_server:5380/api/dhcp/scopes/$action?token=$dst_dns_token&name=$scope" \
    --silent --write-out "%{http_code}" -o response_body.txt)

  echo "HTTP response code: $response"
  if [[ "$response" == "200" ]]; then
    echo "Successfully $action DHCP for scope: $scope"
  else
    echo "Failed to $action DHCP for scope: $scope. Check the response body for details."
  fi

  echo "Response body:"
  cat response_body.txt
done

@besmirzanaj
Copy link
Author

I Like this solution, just need to make sure that the primary DNS server is hosting the zone(s) as "Primary" and the secondary server is configured to host the zone(s) as Secondary. This will need further explanation or tutorials from my original blog post, or re-write it (Which I think would be best)

Docs: https://blog.technitium.com/2022/06/how-to-self-host-your-own-domain-name.html

I tried the Primary, Secondary guide from the article above and it worked flawlessly (I assume it worked for you as well since you proposed the solution :) )

Will have to find the time to properly write up this. Thank you for the contribution

@ShauneBoy
Copy link

ShauneBoy commented Jan 14, 2025

Just what I've been looking for (having just moved from a combination of pi-hole and the DNS server running on my NAS - mainly forced to move as NAS died so thought I would "do it properly") - DNS Syncing is working a treat (having read the above) - setup as Master and Secondary zones but have a couple of questions...

Can you (easily) just replicate the DHCP settings? As think this would complete the setup... The original sync script sync'd everything - whilst the new "swap DHCP server" script just monitors the DHCP - but I have some reservations defined and would like not have to remember to update both DHCP servers ;)... Also, guess it would keep the stand-by DHCP server updated with the current IP addresses in use? [as when I swapped from Pi-Hole, I noted that the clients picked up totally new IP addresses and didn't keep what they had been previously assigned - so would that be the case again if the DHCP server swapped?]

But also wondering if it would be enough to leave the DHCP server enabled on the secondary machine but set the OFFER time to something like 4 seconds? [so the Primary one would normally answer?] - although that doesn't solve the replication of the reservations/current leases :)...

Sorry if any of this seems obvious!

@besmirzanaj
Copy link
Author

Following the updated script, to optimize the DHCP leases, the solution is really simple.

On the Secondary technitium server DHCP settings, configure the "Offer Delay Time" to something higher than 1s (1000 ms), and leave the default 100ms on the Primary one. That means that the Primary Server will always be the first to offer the IP assignment and if it is not available, the second DNS server will take over.

iScreen Shoter - Google Chrome - 250114180425

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment