|
I had two load balancers running slackware 8.0, heartbeat and lvs in a
direct routing topology. I just upgraded one load balancer to RH 8.0
and Ultramonkey 2.0.1 to take advantage of automated package management
systems. Everything seemed to go smoothly and my (now) assymetric load
balancers are talking to each other, but I have an odd problem on the
newly upgraded box. Heartbeat seems to be trying to setup all resources
twice. Here's my haresources file:
lb2 63.77.16.75 10.0.0.249 ldirectord::ldirectord.cf
and here's an excerpt from the ha-debug log file:
heartbeat: 2003/09/25_12:31:40 debug: Starting
/etc/ha.d/resource.d/IPaddr 63.77.16.75 start
heartbeat: 2003/09/25_12:31:40 debug: Starting
/etc/ha.d/resource.d/IPaddr 63.77.16.75 start
ls: /var/lib/heartbeat/rsctmp/IPaddr/eth0:*: No such file or directory
ls: /var/lib/heartbeat/rsctmp/IPaddr/eth0:*: No such file or directory
SIOCSIFBRDADDR: Cannot assign requested address
SIOCSIFFLAGS: Cannot assign requested address
ERROR: /sbin/ifconfig eth0:3 63.77.16.75 netmask 255.255.255.192
broadcast 63.77.16.127 failed.
heartbeat: 2003/09/25_12:31:40 debug: /etc/ha.d/resource.d/IPaddr
63.77.16.75 start done. RC=255
heartbeat: 2003/09/25_12:31:40 debug: /etc/ha.d/resource.d/IPaddr
63.77.16.75 start done. RC=0
heartbeat: 2003/09/25_12:31:40 debug: Starting
/etc/ha.d/resource.d/IPaddr 10.0.0.249 start
heartbeat: 2003/09/25_12:31:40 debug: Starting
/etc/ha.d/resource.d/IPaddr 10.0.0.249 start
heartbeat: 2003/09/25_12:31:41 debug: /etc/ha.d/resource.d/IPaddr
10.0.0.249 start done. RC=0
heartbeat: 2003/09/25_12:31:41 debug: Starting
/etc/ha.d/resource.d/ldirectord ldirectord.cf start
SIOCSIFBRDADDR: Cannot assign requested address
SIOCSIFFLAGS: Cannot assign requested address
ERROR: /sbin/ifconfig eth0:5 10.0.0.249 netmask 255.255.255.192
broadcast 10.0.0.255 failed.
heartbeat: 2003/09/25_12:31:41 debug: /etc/ha.d/resource.d/IPaddr
10.0.0.249 start done. RC=255
heartbeat: 2003/09/25_12:31:41 debug: Starting
/etc/ha.d/resource.d/ldirectord ldirectord.cf start
heartbeat: 2003/09/25_12:31:43 debug: /etc/ha.d/resource.d/ldirectord
ldirectord.cf start done. RC=0
heartbeat: 2003/09/25_12:31:43 debug: /etc/ha.d/resource.d/ldirectord
ldirectord.cf start done. RC=0
You can see that it tries to set each IP twice (one works and one fails,
obviously) and runs ldirectord twice. The biggest problem with this is
that it leaves a rogue ldirectord process running out there.
Ideas? I'm scratching my head on this one...
Thanks!
Kelly
--
--------------------------------------------
-- Kelly Corbin
-- Network Administrator
--
-- http://www.theiqgroup.com
--
-- The IQ Group, Inc.
-- 6740 Antioch Suite 260
-- Merriam, KS 66204
-- (913)-722-6700 x105
-- Fax (913)722-7264
--------------------------------------------
| 