> I've been messing with this recently - OMSHell, not releasing leases - and
IIRC [and I may well not, but take it for what it's worth]...
> you have to expire the lease and then release it.
>
> You can't go from an active lease to released in one step.
>
> I think there are some details in the list that I recall reading recently
- which is why I *vaguely* know. :)
>
> HTH
> -Greg
Thanks for the suggestion! I did try this and I got the following logs:
Jul 30 14:55:52 loc3 dhcpd[25263]: lease 1.1.1.1 state changed from active
to expired
Jul 30 14:55:52 loc3 dhcpd[25263]: bind update on 1.1.1.1 from xxx rejected:
1.1.1.1: invalid state transition: active to expired
Jul 30 14:56:22 loc3 dhcpd[25263]: lease 1.1.1.1 state changed from expired
to released
Jul 30 14:57:13 loc3 dhcpd[25263]: DHCPDISCOVER from aa:bb:cc:dd:ee:ff via
1.1.1.0: network 1.1.1.0/31: no free leases
Jul 30 14:57:16 loc3 dhcpd[25263]: DHCPDISCOVER from aa:bb:cc:dd:ee:ff via
1.1.1.0: network 1.1.1.0/31: no free leases
Jul 30 14:57:19 loc3 dhcpd[25263]: DHCPDISCOVER from aa:bb:cc:dd:ee:ff via
1.1.1.0: network 1.1.1.0/31: no free leases
Jul 30 14:57:22 loc3 dhcpd[25263]: DHCPDISCOVER from aa:bb:cc:dd:ee:ff via
1.1.1.0: network 1.1.1.0/31: no free leases
But I think you may be on the right track, it seems like the previous
guidance of just releasing the lease with OMAPI is not sufficient, maybe
something has changed in newer versions? I was not able to get this process
to work at all on a test server without failover either, so I am thinking
that the failover detail may not be related to the issue.
>
https://en.wikipedia.org/wiki/IPv4#First_and_last_subnet_addresses>
> I'm speaking about the subnet assigned to the NIC.
>
> Bill
See RFC 3021 if you have questions about this, it is pretty widely supported
these days on routers and Linux hosts. The major exception being Windows
hosts, which will not accept this type of subnet mask. Per the Wikipedia
article, "As a special case, a /31 network has capacity for just two hosts.
These networks are typically used for point-to-point connections. There is
no network identifier or broadcast address for these networks."
https://tools.ietf.org/html/rfc3021I am not sure if this is related to the problem or not, but I will try these
same experiments next week with a /30 subnet just in case. It's certainly
possible there could be some kind of bug.
Thanks,
Isaiah
_______________________________________________
ISC funds the development of this software with paid support subscriptions. Contact us at
https://www.isc.org/contact/ for more information.
dhcp-users mailing list
[hidden email]
https://lists.isc.org/mailman/listinfo/dhcp-users