[Pacemaker] HA DHCP3 server on Ubuntu

[Simon Jansen]

Hi,

first of all, thanks for the answers.

> error: Please send logs. Especially those with "ERROR" in it ;-)
Here is the output of the DHCP service start:
Internet Systems Consortium DHCP Server V3.1.3
Copyright 2004-2009 Internet Systems Consortium.
All rights reserved.
For info, please visit https://www.isc.org/software/dhcp/
Wrote 1 leases to leases file.
Listening on LPF/eth2/00:50:56:a0:64:1d/192.168.200/24
Sending on   LPF/eth2/00:50:56:a0:64:1d/192.168.200/24

No subnet declaration for eth1 (192.168.100.10).
** Ignoring requests on eth1.  If this is not what
   you want, please write a subnet declaration
   in your dhcpd.conf file for the network segment
   to which interface eth1 is attached. **


No subnet declaration for eth0 (10.32.194.248).
** Ignoring requests on eth0.  If this is not what
   you want, please write a subnet declaration
   in your dhcpd.conf file for the network segment
   to which interface eth0 is attached. **

Sending on   Socket/fallback/fallback-net


And the entries in the /var/log/messages file.
Oct 27 10:37:31 node dhcpd: Internet Systems Consortium DHCP Server V3.1.3
Oct 27 10:37:31 node dhcpd: Copyright 2004-2009 Internet Systems Consortium.
Oct 27 10:37:31 node dhcpd: All rights reserved.
Oct 27 10:37:31 node dhcpd: For info, please visit
https://www.isc.org/software/dhcp/
Oct 27 10:37:31 node dhcpd: Wrote 1 leases to leases file.
Oct 27 10:37:32 node dhcpd: Listening on
LPF/eth2/00:50:56:a0:64:1d/192.168.200/24
Oct 27 10:37:32 node dhcpd: Sending on
LPF/eth2/00:50:56:a0:64:1d/192.168.200/24
Oct 27 10:37:32 node dhcpd: Sending on   Socket/fallback/fallback-net
Oct 27 10:37:41 node dhcpd: DHCPDISCOVER from 00:50:56:a0:41:16 via eth2
Oct 27 10:37:42 node dhcpd: DHCPOFFER on 192.168.200.101 to
00:50:56:a0:41:16 (node) via eth2
Oct 27 10:37:42 node dhcpd: DHCPREQUEST for 192.168.200.101 (192.168.200.10)
from 00:50:56:a0:41:16 (node) via eth2
Oct 27 10:37:42 node dhcpd: DHCPACK on 192.168.200.101 to 00:50:56:a0:41:16
(node) via eth2
Oct 27 10:38:12 node dhcpd: DHCPREQUEST for 192.168.200.101 from
00:50:56:a0:41:16 (node) via eth2
Oct 27 10:38:12 node dhcpd: DHCPACK on 192.168.200.101 to 00:50:56:a0:41:16
(node) via eth2

Referring to these two logs the service starts fine. But the RA doesn't get
the correct start at all.
Is there a log file where the RA writes debugging or error messages to?


> RA: I'd prefer the init script since it is more specific for this service.

> Please check if a good status() section is included that detects possible
> problems.
Ok. Thanks for the advice. The status section of the init script just ckecks
if the pid file exists and if the ID in the file is equal to the ID in the
output of 'ps -p'. Is there a better way to check if the service is running
correctly or more exactly: is there a command to check if the dhcp server
replies on requests?

Thank you for the fantastic book that you wrote about Pacemaker and
HA-clustering. It's a very good help in many situations. ;)


@Christian: Thanks for the advice. I've seen the builtin failover mechanism
before but with Pacemaker I have a central management interface. So it's
easier to manage. Further we don't want to split the scope.


Regards,

Simon
--
Simon Jansen
64293 Darmstadt
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.clusterlabs.org/pipermail/pacemaker/attachments/20101027/ce1bb58c/attachment-0001.html>