No Failover works, Failover does not work
Shlomo Dubrowin
shlomo at dubrowin.org
Thu May 18 05:05:17 UTC 2006
Thank you all for your help and support. This is now working correctly.
The problem appeard to be:
- The time on the 2 servers were not the same
- It appeared because mclt was configured on both the primary and secondary,
they were both waiting to "recover". As soon as I removed that from the
secondary and restarted the service on the secondary, I was able to get a
lease.
Thank you again for your help.
Shlomo
On 5/18/06, Glenn Satchell <Glenn.Satchell at uniq.com.au> wrote:
>
> Hi Shlomo
>
> I assume that since you mentioned that both servers work ok without
> failover enabled they have valid leases in their dhcpd.leases file. In
> this situation, when failover first starts up the two servers will wait
> for mclt (1800 seconds in your config) to pass so that any temporary
> leases will have expired.
>
> So you have two choices:
>
> 1. Wait for 1800 seconds to pass, and things should magically just
> start to work.
>
> 2. Replace the leases file on the secondary with an empty file. In this
> situation the sceondary wikll transfer the leases over from the primary
> and both will start to issue leases immediately.
>
> The recovering situation occurs when both servers have leases, but
> haven't previously synchronised to each other. Once synchronised they
> will restart immediately.
>
> regards,
> -glenn
>
> >Date: Wed, 17 May 2006 12:35:41 -0600
> >From: Michael Loftis <mloftis at wgops.com>
> >To: dhcp-users at isc.org
> >Subject: Re: No Failover works, Failover does not work
> >Content-Disposition: inline
> >
> >That's DHCPD stating it's in recovery mode. IE *it* is recovering from a
> >failover mode or failure mode or jsut starting up...I *think* it'll clear
> >up on it's own after a little while depending upon your timeout settings.
> >
> >--On May 17, 2006 9:29:38 PM +0300 Shlomo Dubrowin <shlomo at dubrowin.org>
> >wrote:
> >
> >> OK, so I switched the peer ports to 519 and 520.
> >> Now, when I try to get an address, these messages were recorded.
> >> (Franklin is the primary DHCP Server, the MAC Address listed is my
> >> machine, and the IP 10.153.22.1 is the router IP for the VLAN that is
> >> configured as the relay agent.):*
> >>
> >> *messages:May 17 21:18:37 franklin dhcpd: DHCPDISCOVER from
> >> 00:12:3f:1a:60:c3 via 10.153.22.1: not responding (recovering)
> >> messages:May 17 21:18:42 franklin dhcpd: DHCPDISCOVER from
> >> 00:12:3f:1a:60:c3 via 10.153.22.1: not responding (recovering)
> >> messages:May 17 21:18:50 franklin dhcpd: DHCPDISCOVER from
> >> 00:12:3f:1a:60:c3 via 10.153.22.1: not responding (recovering)
> >> messages:May 17 21:19:06 franklin dhcpd: DHCPDISCOVER from
> >> 00:12:3f:1a:60:c3 via 10.153.22.1: not responding (recovering)
> >>
> >> So which machine isn't responding? the machine trying to get an
> address
> >> (the MAC Address) or the cisco router with the IP?
> >>
> >> Shlomo
> >>
> >> On 5/17/06, Shlomo Dubrowin <shlomo at dubrowin.org> wrote:
> >>>
> >>> OK, I can try that, the documentation I was following said they could
> be
> >>> the same. thanx.
> >>>
> >>> Shlomo
> >>>
> >>>
> >>> On 5/17/06, Karen Pachla < kpachla at umich.edu> wrote:
> >>> >
> >>> > Hi,
> >>> >
> >>> > On my configuration, I have primary's port at 519, secondary as 520
> >>> >
> >>> > Primary:
> >>> > port 519;
> >>> > peer port 520;
> >>> >
> >>> > Secondary:
> >>> > port 520;
> >>> > peer port 519;
> >>> >
> >>> >
> >>> > On Wed, 2006-05-17 at 20:51 +0300, Shlomo Dubrowin wrote:
> >>> >
> >>> > > Primary Config:
> >>> > > -----------------------
> >>> > > failover peer "dhcp-failover" {
> >>> > > primary; # declare this to be the primary server
> >>> > > address 10.153.100.26;
> >>> > > port 520;
> >>> > > peer address 10.153.100.27;
> >>> > > peer port 520;
> >>> > > max-response-delay 30;
> >>> > > max-unacked-updates 30;
> >>> > > load balance max seconds 3;
> >>> > > mclt 1800;
> >>> > > split 128;
> >>> > > }
> >>> > >
> >>> > > include "/etc/dhcpd.pools";
> >>> > >
> >>> > >
> >>> > > Secondary Config:
> >>> > > ----------------------------
> >>> > > failover peer "dhcp-failover" {
> >>> > > Secondary; # declare this to be the secondary server
> >>> > > address 10.153.100.27;
> >>> > > port 520;
> >>> > > peer address 10.153.100.26;
> >>> > > peer port 520;
> >>> > > max-response-delay 30;
> >>> > > max-unacked-updates 30;
> >>> > > load balance max seconds 3;
> >>> > > mclt 1800;
> >>> > > }
> >>> >
> >>> > --
> >>> > K. Pachla
> >>> > LSAIT
> >>> > University of Michigan
> >>> >
> >>> >
> >>> >
> >>>
> >>>
> >>> --
> >>>
> -----------------------------------------------------------------------
> >>>
> >>> ,-~~-.___. ._.
> >>> / | ' \ | |"""""""""| Shlomo Dubrowin
> >>> ( ) 0 | | | (Sheldon)
> >>> \_/-, ,----' | | |
> >>> ==== !_!--v---v--"
> >>> / \-'~; |""""""""| shlomo at dubrowin.org
> >>> / __/~| ._-""|| | http://www.dubrowin.org
> >>> =( _____|_|____||________|
> >>>
> -----------------------------------------------------------------------
> >>>
> >>
> >>
> >>
> >> --
> >> -----------------------------------------------------------------------
> >> ,-~~-.___. ._.
> >> / | ' \ | |"""""""""| Shlomo Dubrowin
> >> ( ) 0 | | | (Sheldon)
> >> \_/-, ,----' | | |
> >> ==== !_!--v---v--"
> >> / \-'~; |""""""""| shlomo at dubrowin.org
> >> / __/~| ._-""|| | http://www.dubrowin.org
> >> =( _____|_|____||________|
> >> -----------------------------------------------------------------------
> >>
> >>
> >>
> >
> >
> >
> >--
> >"Genius might be described as a supreme capacity for getting its
> possessors
> >into trouble of all kinds."
> >-- Samuel Butler
> >
>
>
>
--
-----------------------------------------------------------------------
,-~~-.___. ._.
/ | ' \ | |"""""""""| Shlomo Dubrowin
( ) 0 | | | (Sheldon)
\_/-, ,----' | | |
==== !_!--v---v--"
/ \-'~; |""""""""| shlomo at dubrowin.org
/ __/~| ._-""|| | http://www.dubrowin.org
=( _____|_|____||________|
-----------------------------------------------------------------------
More information about the dhcp-users
mailing list