DHCP failover won't start
Glenn Satchell
Glenn.Satchell at uniq.com.au
Tue Sep 4 13:03:18 UTC 2007
Here's what I do for syslog in dhcpd.conf (if local7 already used pick
one from local0 to local6):
log-facility local7;
and in /etc/syslog.conf:
# DHCP
local7.debug /var/log/dhcp/dhcpd.log
Note whitespace in syslog.conf *must* be tabs in Solaris. Then create
/var/log/dhcp/dhcpd.log:
mkdir -p /var/log/dhcp
touch /var/log/dhcp/dhcpd.log
svcadm -v restart svc:/system/system-log:default
Now stop dhcpd on both, create zero length dhcpd.conf file on both.
Start dhcpd on secondary, then primary. Look in /var/log/dhcp/dhcpd.log
for messages. If it is still not working, post both the log files please.
regards,
-glenn
>From: "Arno _" <r_no at hotmail.com>
>To: dhcp-users at isc.org
>Subject: Re: DHCP failover won't start
>Date: Tue, 04 Sep 2007 14:47:54 +0200
>
>When Editing the lease-file.db I first stop both dhcpd service, so it' not
>suppose to be confused !
>And I can wait for a long time without event (more than 1800 second or
>minutes)
>
>on /var/adm/messages I have nothing.
>
>So I change the port number to have to different one: 519 and 520, as
>someone suggest.
>
>
>So it's an hour after the launch of the 2 dhcpd, still in recover,
>unknown-state !!
>
>
>>From: Glenn Satchell <Glenn.Satchell at uniq.com.au>
>>Reply-To: dhcp-users at isc.org
>>To: dhcp-users at isc.org
>>Subject: Re: DHCP failover won't start
>>Date: Tue, 4 Sep 2007 22:17:22 +1000 (EST)
>>
>>Ahh. Recover mode is used when the hosts have communicated before - it
>>will wait until mtbf expires (1800 seconds) before moving to the next
>>step. If these are new servers, try creating zero length dhcpd.conf
>>files and restarting. It's possible that your editting of the
>>dhcpd.leases file has confused the daemon. See the dhcpd.conf man page,
>>the section on FAILOVER STARTUP.
>>
>>Also look in /var/adm/messages, you will get some dhcp messages there,
>>including status after dhcpd starts:
>>
>>Sep 4 22:08:02 chisel dhcpd: [ID 702911 local7.info] failover peer
>>Uniq14subnet: peer moves from normal to communications-interrupted
>>Sep 4 22:08:02 chisel dhcpd: [ID 702911 local7.info] failover peer
>>Uniq14subnet: I move from startup to normal
>>
>>Or you may need to log daemon.debug to a separate file to get the full
>>range of messages.
>>
>>regards,
>>-glenn
>>
>> >From: "Arno _" <r_no at hotmail.com>
>> >To: dhcp-users at isc.org
>> >Subject: Re: DHCP failover won't start
>> >Date: Tue, 04 Sep 2007 13:53:14 +0200
>> >
>> >here it is:
>> >master config:
>> >failover peer "dhcp-failover" {
>> > primary;
>> > address 172.24.1.5;
>> > port 520;
>> > peer address 172.24.1.6;
>> > peer port 520;
>> > max-response-delay 60;
>> > max-unacked-updates 10;
>> > mclt 1800;
>> > split 128;
>> > load balance max seconds 2;
>> >}
>> >
>> >ns1:/opt/dhcpd # /etc/init.d/dhcpd start
>> >starting dhcp deamon
>> >Internet Systems Consortium DHCP Server V3.0.5
>> >Copyright 2004-2006 Internet Systems Consortium.
>> >All rights reserved.
>> >For info, please visit http://www.isc.org/sw/dhcp/
>> >Wrote 0 deleted host decls to leases file.
>> >Wrote 0 new dynamic host decls to leases file.
>> >Wrote 0 leases to leases file.
>> >Listening on DLPI/nge0/00:e0:81:5e:47:c4/172.24.1/24
>> >Sending on DLPI/nge0/00:e0:81:5e:47:c4/172.24.1/24
>> >Sending on Socket/fallback/fallback-net
>> >failover peer dhcp-failover: I move from recover to startup
>> >
>> >lease-file.db:
>> >failover peer "dhcp-failover" state {
>> > my state recover at 2 2007/09/04 11:44:16;
>> > partner state unknown-state at 2 2007/09/04 11:44:16;
>> >}
>> >
>> >
>> >secondary config:
>> >failover peer "dhcp-failover" {
>> > secondary;
>> > address 172.24.1.6;
>> > port 520;
>> > peer address 172.24.1.5;
>> > peer port 520;
>> > max-response-delay 60;
>> > max-unacked-updates 10;
>> > load balance max seconds 2;
>> >}
>> >
>> >starting dhcp deamon
>> >Internet Systems Consortium DHCP Server V3.1.0
>> >Copyright 2004-2007 Internet Systems Consortium.
>> >All rights reserved.
>> >For info, please visit http://www.isc.org/sw/dhcp/
>> >Wrote 0 deleted host decls to leases file.
>> >Wrote 0 new dynamic host decls to leases file.
>> >Wrote 0 leases to leases file.
>> >Listening on DLPI/nge0/00:e0:81:71:9b:fe/172.24.1/24
>> >Sending on DLPI/nge0/00:e0:81:71:9b:fe/172.24.1/24
>> >Sending on Socket/fallback/fallback-net
>> >failover peer dhcp-failover: I move from recover to startup
>> >
>> >lease-file.db:
>> >failover peer "dhcp-failover" state {
>> > my state recover at 2 2007/09/04 11:44:46;
>> > partner state unknown-state at 2 2007/09/04 11:44:46;
>> > mclt 0;
>> >}
>> >
>> >
>> >and it's running on solaris 10 X86, I can see some traffic with snoop
>> >between the 2 server, but 1 or 2 packet.
>> >
>> >thanks for your help,
>> >
>> >
>> >>From: Glenn Satchell <Glenn.Satchell at uniq.com.au>
>> >>Reply-To: dhcp-users at isc.org
>> >>To: dhcp-users at isc.org
>> >>Subject: Re: DHCP failover won't start
>> >>Date: Tue, 4 Sep 2007 21:38:28 +1000 (EST)
>> >>
>> >>
>> >> >X-Originating-Email: [r_no at hotmail.com]
>> >> >From: "Arno _" <r_no at hotmail.com>
>> >> >To: dhcp-users at isc.org
>> >> >Subject: DHCP failover won't start
>> >> >Date: Tue, 04 Sep 2007 13:04:46 +0200
>> >> >
>> >> >Hello,
>> >> >I have installed the DHCP 3.1.0 on my solaris box, but the failover
>>won't
>> >> >start, it always show me:
>> >> >failover peer "dhcp-failover" state {
>> >> >my state recover at 2 2007/09/04 10:52:51;
>> >> >partner state unkown-state at 2 2007/09/04 10:52:51;
>> >> >}
>> >> >
>> >> >same thing on the secondary. and won't change in it's state.
>> >> >I try to setup the lease-file with something like:
>> >> >failover peer "dhcp-failover" state {
>> >> >my state partner-down;
>> >> >partner state unkown-state at 2 2007/09/04 10:52:51;
>> >> >}
>> >> >
>> >> >like it was explain in the man of the dhcpd.conf, but that won't
>>change
>> >> >anything.
>> >> >
>> >> >What can I do ?
>> >> >
>> >> >I know the config is good because I have on other dhcp server, and
>>also I
>> >> >know I have to setup the lease-file in a certzain way for the first
>> >>start,
>> >> >but can't find a clear way to do it
>> >> >
>> >> >thanks for your help.
>> >>
>> >>Please post your dhcp configuration, especially the bit where "failover
>> >>peer" is defined for both servers (as it should be different). Post the
>> >>startup messages that get wrtten out to the terminal when dhcpd starts
>> >>on each server. The mailing list software does not accept attachemnts
>> >>so paste it in the body of the email.
>> >>
>> >>The rest here is a guess, but things to check:
>> >>
>> >>The time on the two servers is pretty close. Best bet i sto use NTP to
>> >>synchronise the clocks accurately.
>> >>
>> >>That you can ping the other server. If there is a firewall in between
>> >>check that communication is allowed on the ports specified in the
>> >>failover peer definition.
>> >>
>> >>You do not need to put anything special in dhcpd.leases. In fact for a
>> >>new server this should be a zero length file. dhcpd will write the
>> >>status information once it connects to its peer.
>> >>
>> >>regards,
>> >>-glenn
>> >>
>> >
>> >_________________________________________________________________
>> >FREE pop-up blocking with the new MSN Toolbar - get it now!
>> >http://toolbar.msn.click-url.com/go/onm00200415ave/direct/01/
>> >
>> >
>>
>
>_________________________________________________________________
>Test your celebrity IQ. Play Red Carpet Reveal and earn great prizes!
>http://club.live.com/red_carpet_reveal.aspx?icid=redcarpet_hotmailtextlink2
>
>
More information about the dhcp-users
mailing list