Closed sargonphin closed 11 months ago
Debug token URL as requested in the template please.
This is expected and should also be documented somewhere: Static DHCP leases are configured using dhcp-host
config options for the dnsmasq
server which re-reads its full config only during a restart (some parts like changing /etc/hosts
files can be done at runtime, but this is rather an exception). As you reported that restarting pihole-FTL
solves the issue, I truly think there is no issue at even begin with. The "workaround" you describe is what is really happening, expected and designed to work like this.
I'm going to close this as I see no bug here. If you still think there is one, feel free to comment + reopen at any point.
Hello! Really really sorry I didn't add the debug link and took so much time to reply... I was super duper busy!
I understand the inner workings now and why it is not behaving as I would like. I suggest to prompt to restart the FTL service on the configuration page every time a static lease is added, or at least have a warning stating that the changes will only take effect upon restarting the service.
It is not mentionned anywhere clearly, sadly :( I think this could be improved
Moving as this is a question for the web interface.
Currently working on a note to the web interface for this.
Really cool, thank you a lot :) A button to restart the service directly from the UI would be the best, is it possible?
Go to Settings > System > Restart Resolver
Thanks for the tip :D If that's possible, could it be added as a direct access button when changing DHCP settings? In the same place as the warning about the DHCP changes?
No, I wouldn't do that. It fits well in System and there are other reasons as to restart the resolver than DHCP. I think you can effort an additional click.
Oh I did not meant to move it to the DHCP settings! What I meant was to put a link/button to restart the resolver on the same warning window that will say that DHCP changes requires a resolver restart :) basically, a duplicate ^_^
This issue is stale because it has been open 30 days with no activity. Please comment or update this issue or it will be closed in 5 days.
Hey, can we reopen this issue, please?
I've run into this a couple times now and it's been quite frustrating. The note suggested would go long way to help myself and other users avoid this confusion in the future. :)
This issue is stale because it has been open 30 days with no activity. Please comment or update this issue or it will be closed in 5 days.
Versions
Platform
Expected behavior
Following a new connection from a new device, the new device gets an address from the DHCP pool. In Pi Hole Web configuration panel, after assigning a static lease, a DHCP renew on the new device (release + request) should fail with a NAK response from the DHCP server, offering the address defined in the static lease instead.
Actual behavior / bug
The server bypasses the static leases, uses existing leases saved in database and acknowledges the request from the new device, which ends up getting the same DHCP pool address as before instead of the static lease IP.
In Pi Hole Web configuration panel, deleting the lease (before or after disconnecting the new device) makes no difference.
Restarting the PiHole FTL service seems to be a workaround. The issue fixes itself after the initial DHCP pool lease has expired.
Steps to reproduce
Steps to reproduce the behavior: