Failover pair in strange state

Eustace, Glen G.Eustace at massey.ac.nz
Mon Jun 8 20:33:34 UTC 2009


Turned out the issue was that ntp was no longer synchronising time on one server ( no idea yet why ) and the time had crpt forward slowly to be approx 90secs ahead of the other.  Result: failover protocol no longer works.  Fix: stop ntpd, sync time, restart ntpd restart dhcpd on one server.

From: dhcp-users-bounces at lists.isc.org [mailto:dhcp-users-bounces at lists.isc.org] On Behalf Of Eustace, Glen
Sent: Monday, June 08, 2009 10:03 AM
To: dhcp-users at lists.isc.org
Subject: Failover pair in strange state

We have two servers that have been working fine in a failover pair for a while now.

Now , however, there is no pool balancing activity occurring at all.  Looking the leases file indicates that both think they are not healthy.

Server wel-net1:

failover peer "wel-dhcp-failover" state {
  my state communications-interrupted at 4 2009/06/04 03:00:55;
  partner state normal at 1 2009/05/11 23:24:07;
}

Server wel-net2:

failover peer "wel-dhcp-failover" state {
  my state communications-interrupted at 4 2009/06/04 03:03:37;
  partner state normal at 1 2009/05/11 22:10:48;
  mclt 1800;
}

Shutting down the dhcp processes doesn't seem to make any difference, any clues ?  How do we get the peering relationship re-established. To the best of our knowledge no changes have been made in the environment.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.isc.org/pipermail/dhcp-users/attachments/20090609/540cef26/attachment.html>


More information about the dhcp-users mailing list