<html><head><title>Re: Failback causes lost lease</title>
<META http-equiv=Content-Type content="text/html; charset=windows-1252">
</head>
<body>
<br><br>
<table bgcolor="#ffffff">
<tr>
<td width=2 bgcolor= #0000ff><br>
</td>
<td><span style=" font-family:'courier new'; font-size: 9pt;">On 06/25/2015 01:39 PM, Gregory Sloop wrote:<br>
Re: Failback causes lost lease <br>
</span><table bgcolor="#ffffff">
<tr>
<td width=2 bgcolor= #0000ff><br>
</td>
<td><span style=" font-family:'courier new'; font-size: 9pt;">Gregory,<br>
<br>
Thanks for your reply.<br>
<br>
On 06/25/2015 12:47 PM, Gregory Sloop wrote:<br>
Re: Failback causes lost lease <span style=" color: #800000;"><b>SM> In testing my dhcp failover, I pulled the ethernet cable on the primary<br>
SM> server.<br>
<br>
SM> The secondary server acknowleged renewal requests as expected.<br>
<br>
SM> Then I plugged the cable back in. After both the primary and secondary<br>
SM> had moved from communications-interrupted to normal, the secondary logs<br>
SM> multiple dhcp requests from a client whose lease is owned by the primary<br>
SM> server. The primary server does not log any of these but the last <br>
SM> request, reporting that "lease in transition state expired".<br>
<br>
SM> Then the secondary server logs a DHCPDISCOVER from that client and <br>
SM> records it load balancing to the primary server.<br>
<br>
SM> The primary server also sees the DHCPDISCOVER and offers a new lease <br>
SM> that is not the same number as the previous lease. This despite the old<br>
SM> number not having been reassigned.<br>
<br>
SM> The end result is that failback causes my clients to change their ip <br>
SM> address.<br>
<br>
SM> Why does this happen and how can I prevent it?<br>
<br>
SM> _______________________________________________<br>
SM> dhcp-users mailing list<br>
</b></span></span><a style=" font-family:'courier new'; font-size: 9pt;" href="mailto:dhcp-users@lists.isc.org">SM> dhcp-users@lists.isc.org</a><br>
<a style=" font-family:'courier new'; font-size: 9pt;" href="https://lists.isc.org/mailman/listinfo/dhcp-users">SM> https://lists.isc.org/mailman/listinfo/dhcp-users</a><br>
<br>
<span style=" font-family:'courier new'; font-size: 9pt;">1) Logs would be good.</td>
</tr>
</table>
<span style=" font-family:'courier new'; font-size: 9pt;">Primary Server:<br>
=========<br>
<br>
Jun 25 11:33:00 dhcpd: DHCPREQUEST for 10.112.1.1 from 00:19:d1:4d:03:f8 via em2.112<br>
Jun 25 11:33:00 dhcpd: DHCPACK on 10.112.1.1 to 00:19:d1:4d:03:f8 via em2.112<br>
Jun 25 11:36:23 dhcpd: timeout waiting for failover peer partner<br>
Jun 25 11:36:23 dhcpd: peer partner: disconnected<br>
Jun 25 11:36:23 dhcpd: failover peer partner: I move from normal to communications-interrupted<br>
Jun 25 11:42:38 dhcpd: failover peer partner: peer moves from normal to communications-interrupted<br>
Jun 25 11:42:38 dhcpd: failover peer partner: I move from communications-interrupted to normal<br>
Jun 25 11:42:38 dhcpd: balancing pool 7f4c9a14cb80 10.112.0.0/13 total 1024 free 511 backup 512 lts 0 max-own (+/-)102<br>
Jun 25 11:42:38 dhcpd: balanced pool 7f4c9a14cb80 10.112.0.0/13 total 1024 free 511 backup 512 lts 0 max-misbal 153<br>
Jun 25 11:42:38 dhcpd: balancing pool 7f4c9a143260 10.104.0.0/13 total 1024 free 512 backup 512 lts 0 max-own (+/-)102<br>
Jun 25 11:42:38 dhcpd: balanced pool 7f4c9a143260 10.104.0.0/13 total 1024 free 512 backup 512 lts 0 max-misbal 154<br>
Jun 25 11:42:38 dhcpd: balancing pool 7f4c9a13a930 10.96.0.0/13 total 1024 free 512 backup 512 lts 0 max-own (+/-)102<br>
Jun 25 11:42:38 dhcpd: balanced pool 7f4c9a13a930 10.96.0.0/13 total 1024 free 512 backup 512 lts 0 max-misbal 154<br>
Jun 25 11:42:38 dhcpd: failover peer partner: peer moves from communications-interrupted to normal<br>
Jun 25 11:43:38 dhcpd: DHCPREQUEST for 10.112.1.1 from 00:19:d1:4d:03:f8 via em2.112: lease in transition state expired<br>
Jun 25 11:43:39 dhcpd: DHCPDISCOVER from 00:19:d1:4d:03:f8 via em2.112<br>
Jun 25 11:43:40 dhcpd: DHCPOFFER on 10.112.2.201 to 00:19:d1:4d:03:f8 via em2.112<br>
Jun 25 11:43:40 dhcpd: uid lease 10.112.1.1 for client 00:19:d1:4d:03:f8 is duplicate on 10.112.0.0/13<br>
Jun 25 11:43:40 dhcpd: DHCPREQUEST for 10.112.2.201 (10.112.0.104) from 00:19:d1:4d:03:f8 via em2.112<br>
Jun 25 11:43:40 dhcpd: DHCPACK on 10.112.2.201 to 00:19:d1:4d:03:f8 via em2.112<br>
<br>
Secondary Server:<br>
===========<br>
Jun 25 11:36:21 dhcpd: timeout waiting for failover peer partner<br>
Jun 25 11:36:21 dhcpd: peer partner: disconnected<br>
Jun 25 11:36:21 dhcpd: failover peer partner: I move from normal to communications-interrupted<br>
Jun 25 11:37:34 dhcpd: DHCPREQUEST for 10.112.1.1 from 00:19:d1:4d:03:f8 via em2.112<br>
Jun 25 11:37:34 dhcpd: DHCPACK on 10.112.1.1 to 00:19:d1:4d:03:f8 via em2.112<br>
Jun 25 11:39:10 dhcpd: DHCPREQUEST for 10.112.1.1 from 00:19:d1:4d:03:f8 via em2.112<br>
Jun 25 11:39:10 dhcpd: DHCPACK on 10.112.1.1 to 00:19:d1:4d:03:f8 via em2.112<br>
Jun 25 11:40:11 dhcpd: DHCPREQUEST for 10.112.1.1 from 00:19:d1:4d:03:f8 via em2.112<br>
Jun 25 11:40:11 dhcpd: DHCPACK on 10.112.1.1 to 00:19:d1:4d:03:f8 via em2.112<br>
Jun 25 11:40:44 dhcpd: DHCPREQUEST for 10.112.1.1 from 00:19:d1:4d:03:f8 via em2.112<br>
Jun 25 11:40:44 dhcpd: DHCPACK on 10.112.1.1 to 00:19:d1:4d:03:f8 via em2.112<br>
Jun 25 11:41:16 dhcpd: DHCPREQUEST for 10.112.1.1 from 00:19:d1:4d:03:f8 via em2.112<br>
Jun 25 11:41:16 dhcpd: DHCPACK on 10.112.1.1 to 00:19:d1:4d:03:f8 via em2.112<br>
Jun 25 11:41:47 dhcpd: DHCPREQUEST for 10.112.1.1 from 00:19:d1:4d:03:f8 via em2.112<br>
Jun 25 11:41:47 dhcpd: DHCPACK on 10.112.1.1 to 00:19:d1:4d:03:f8 via em2.112<br>
Jun 25 11:42:12 dhcpd: DHCPREQUEST for 10.112.1.1 from 00:19:d1:4d:03:f8 via em2.112<br>
Jun 25 11:42:12 dhcpd: DHCPACK on 10.112.1.1 to 00:19:d1:4d:03:f8 via em2.112<br>
Jun 25 11:42:37 dhcpd: DHCPREQUEST for 10.112.1.1 from 00:19:d1:4d:03:f8 via em2.112<br>
Jun 25 11:42:37 dhcpd: DHCPACK on 10.112.1.1 to 00:19:d1:4d:03:f8 via em2.112<br>
Jun 25 11:42:38 dhcpd: failover peer partner: peer moves from normal to communications-interrupted<br>
Jun 25 11:42:38 dhcpd: failover peer partner: I move from communications-interrupted to normal<br>
Jun 25 11:42:38 dhcpd: balancing pool 7f325dbfdb30 10.112.0.0/13 total 1024 free 511 backup 512 lts 0 max-own (+/-)102<br>
Jun 25 11:42:38 dhcpd: balanced pool 7f325dbfdb30 10.112.0.0/13 total 1024 free 511 backup 512 lts 0 max-misbal 153<br>
Jun 25 11:42:38 dhcpd: balancing pool 7f325dbf4210 10.104.0.0/13 total 1024 free 512 backup 512 lts 0 max-own (+/-)102<br>
Jun 25 11:42:38 dhcpd: balanced pool 7f325dbf4210 10.104.0.0/13 total 1024 free 512 backup 512 lts 0 max-misbal 154<br>
Jun 25 11:42:38 dhcpd: balancing pool 7f325dbeb8e0 10.96.0.0/13 total 1024 free 512 backup 512 lts 0 max-own (+/-)102<br>
Jun 25 11:42:38 dhcpd: balanced pool 7f325dbeb8e0 10.96.0.0/13 total 1024 free 512 backup 512 lts 0 max-misbal 154<br>
Jun 25 11:42:38 dhcpd: failover peer partner: peer moves from communications-interrupted to normal<br>
Jun 25 11:43:06 dhcpd: DHCPREQUEST for 10.112.1.1 from 00:19:d1:4d:03:f8 via em2.112: lease owned by peer<br>
Jun 25 11:43:13 dhcpd: DHCPREQUEST for 10.112.1.1 from 00:19:d1:4d:03:f8 via em2.112: lease owned by peer<br>
Jun 25 11:43:24 dhcpd: DHCPREQUEST for 10.112.1.1 from 00:19:d1:4d:03:f8 via em2.112: lease owned by peer<br>
Jun 25 11:43:38 dhcpd: DHCPREQUEST for 10.112.1.1 from 00:19:d1:4d:03:f8 via em2.112: lease owned by peer<br>
Jun 25 11:43:39 dhcpd: DHCPDISCOVER from 00:19:d1:4d:03:f8 via em2.112: load balance to peer partner<br>
Jun 25 11:43:40 dhcpd: DHCPREQUEST for 10.112.2.201 (10.112.0.104) from 00:19:d1:4d:03:f8 via em2.112: lease owned by peer<br>
<br>
</span><table bgcolor="#ffffff">
<tr>
<td><span style=" font-family:'courier new'; font-size: 9pt;">2) I think something with your config is broken. If I were to [wildly] guess, it's a physical/network layer issue.<br>
3) I have a very small setup with 100+ clients, and it certainly doesn't work this way for me. <br>
<br>
There are some issues when a single server is up and in "communications interrupted" mode and you've got a tight IP pool and the leases were fairly evenly balanced against both servers. [I've posted, in the past, about an event that was kinda ugly for this client while running a 4.1 version [IIRC]. *However* those problems should be vastly less of a problem with 4.2+ - and you're not having an issue with communications interrupted anyway.</td>
</tr>
</table>
<br><br>
<table bgcolor="#ffffff">
<tr>
<td><span style=" font-family:'courier new'; font-size: 9pt;">I am having an issue with communications interrupted. When I pull the ethernet cable, both the primary and secondary servers move from normal to communications-interrupted.</td>
</tr>
</table>
<span style=" font-family:'courier new'; font-size: 9pt;">But in your initial post on this thread you said: <br>
<br>
> "<span style=" color: #800000;"><b>After both the primary and secondary<br>
> had moved from communications-interrupted to normal"<br>
<br>
</b><span style=" color: #000000;">It can't be both ways. Either they are CI, or in a Normal state. It can't be both.<br>
Like I said, logs would probably be helpful. [Unless someone else has a lightening bolt moment and can tell you exactly what's wrong without them - but I doubt that.<br>
I hope the logs above help.</td>
</tr>
</table>
<br><br>
<br>
<span style=" font-family:'Courier New'; font-size: 9pt;">---<br>
You didn't answer the question: "It can't be both ways. Either they are CI, or in a Normal state."<br>
<br>
But the logs you posted seem to indicate that both servers believe they're in "normal" mode when the DISCOVER requests come in. Yet they only show the primary getting a single DISCOVER [the last one] while the secondary gets (4) over more than 30s.<br>
<br>
Does a packet capture show the primary server actually receiving all the DHCPREQUEST packets that are going to the secondary?<br>
It would be odd [typically a sign of lost packets] that the secondary would see (4) DHCPREQUEST's but the primary would only see one. It would be very helpful to know if those packets are actually arriving at the interface, but being ignored by dhcpd, or if they're not ever getting there at all. If it's the latter, it's probably not a dhcpd problem.<br>
<br>
Searches on "lease in transition state expired" appear to find incidences of that output, as a result of bugs, but I could only find those kinds of problems recounted in *really* old, non-supported versions. [i.e. 3.0.x]<br>
<br>
Are you running a currently supported version, and what version is it?<br>
<br>
<br>
-Greg<br>
</body></html>