F23 does not renew dhcp leases correctly and corrupts DDNS

[Dr J Austin]

On Tue, 2016-01-26 at 14:05 +1030, Tim wrote:
> Allegedly, on or about 25 January 2016, Greg Woods sent:
> > (I can't remember how to get a dump of the zone, but I remember doing
> > it in the past.
> 
> Simply stopping the nameserver ought to cause it to reconcile its
> records on file.  That's what I do when I've struck a DNS/DHCP foul-up.
> Stop DHCP server, Stop BIND, edit records, increment the serial number,
> restart BIND, restart DHCP server.
> 
> That sort of thing tends to happen when you're experimenting, and
> haven't invented a new IP outside of the DHCP pool.
> 
> e.g. If your LAN uses 192.168.0.0, then set aside a range to be dynamic,
> such as 192.168.0.1 to 192.168.0.100, and configure your DHCP server to
> only dole out those addresses to dynamic clients.  For static addresses,
> whether individually set on the client equipment, or handed out as fixed
> DHCP addresses, use IPs outside of that range.
> 
> Other things that confuse DHCP/DNS server combinations are dual-boot
> PCs, where on one OS the DHCP client sends out prefix codes that the
> other does not, and the DHCP server looks at those codes as well as MAC
> addresses, and doesn't give the same PC the same address for each OS.
> Which is logical enough, except that it doesn't properly reset the one
> it'd previously doled out to the same PC.  It gets truly messy if the PC
> uses the same hostname on both OSs.  I've seen reverse IP records stay
> stale, and forward ones updated.
> 
> 
> -- 
> [tim at localhost ~]$ uname -rsvp
> Linux 3.9.10-100.fc17.x86_64 #1 SMP Sun Jul 14 01:31:27 UTC 2013 x86_64
> 
> Boilerplate:  All mail to my mailbox is automatically deleted, there is
> no point trying to privately email me, I only get to see the messages
> posted to the mailing list.
> 
> I'd just like to say that vinyl record crackles and pops are far less
> annoying than digigigigital mu-u-u-u-usic hiccicicicups and
> yooo-----------------u tu-----be ....... pauses.
> 
> 
> 
Thank you for the replies

I am working on a small network and hence I can reset the dhcp/ddns
to "Zero" - I have a script that does this.

I added my experiences to
https://bugzilla.redhat.com/show_bug.cgi?id=1269093
and received a very prompt reply from Dan Williams

I have tested out the use of the two commands
nmcli con mod eno1 connection.autoconnect no
nmcli con mod eno1 connection.autoconnect yes
and by magic the client no longer issues DHCPDISCOVER
every time NetworkManager is restarted and hence does not
request and obtain a new IP address.

I also checked what happened when the leases ran out naturally by
setting "default-lease-time  300;" in
dhcpd.conf on the server

Again all was well

I have not yet worked out how often the magic incantation
is required - once after a new installation or once
before the lease runs out or ...

Aside:
I suspect that originally the client may not have been
deleting the old IP address from the interface but was adding
the new IP as a "secondary".
At one stage during testing "ip a s" gave multiple entries
of the form
inet 148.197.29.133/24 brd 148.197.29.255 scope global dynamic eno1		


This might explain my original symptoms - after several days the client
and server would both freeze with the network busily flashing away.
The only way out was to press the button on all machines.
I originally thought it was a hardware problem but now suspect
that NetworkManager may have been the cause of the crashes.
At least my machines, router, switches, cables, ...
have all been cleaned and dusted!

A nasty little problem as I have been happily running my
arrangement for 10 years or more!