[Linux-HA] Problems with an active/passive apache ha cluster

[Fernando M. Morgenstern]

Hi,

I'm trying to setup an Apache ha cluster, but i'm having some problems 
(which i describe below). It is a simple conf, 192.168.0.1 is the first 
node (which should work as active) and 192.168.0.10 is the second node 
(which should work as passive). Heartbeat2 (according to logs) is 
already working. Well, here are my problems:

1) Even using "auto_failback off" on both nodes ha.cf, they work as 
active/active. But they should work as active/passive right?

2) It seens that when i stop apache on server 192.168.0.1, the other 
node stops to. I can still ping 192.168.0.105 (the virtual address) but 
the web server stop responding. When i start again apache on 
192.168.0.1, everything goes back to normal. Does this happen because of 
unwanted active/active state?

Here are my conf files:

====== server1 ======

==ha.cf==

keepalive 1 # 1 second
deadtime 5
logfile        /var/log/ha-log
logfacility        local0
bcast        eth0                # Linux
mcast eth0 225.0.0.1 694 1 0
auto_failback off
node        c4
node        mail.tifacil.com.br
respawn hacluster /usr/lib/heartbeat/ipfail
apiauth ipfail gid=haclient uid=hacluster

==haresources==

c4        \
        ldirectord::ldirectord.cf \
        LVSSyncDaemonSwap::master \
        IPaddr2::192.168.0.105/24/eth0/192.168.0.255

==ldirectord.cf==

checktimeout=10
checkinterval=2
autoreload=no
logfile="local0"
quiescent=yes

virtual=192.168.0.105:80
        real=192.168.0.1:80 gate
        real=192.168.0.10:80 gate
        fallback=127.0.0.1:80 gate
        service=http
        request="ldirector.html"
        receive="Test Page"
        scheduler=rr
        protocol=tcp
        checktype=negotiate

====== server2 ======

==ha.cf==

keepalive 1 # 1 second
deadtime 5
logfile        /var/log/ha-log
logfacility        local0
bcast        eth0                # Linux
mcast eth0 225.0.0.1 694 1 0
auto_failback off
node        c4
node        mail.tifacil.com.br
respawn hacluster /usr/lib/heartbeat/ipfail
apiauth ipfail gid=haclient uid=hacluster

==haresources==

c4        \
        ldirectord::ldirectord.cf \
        LVSSyncDaemonSwap::master \
        IPaddr2::192.168.0.105/24/eth0/192.168.0.255

==ldirectord.cf==

checktimeout=10
checkinterval=2
autoreload=no
logfile="local0"
quiescent=yes

virtual=192.168.0.105:80
        real=192.168.0.1:80 gate
        real=192.168.0.10:80 gate
        fallback=127.0.0.1:80 gate
        service=http
        request="ldirector.html"
        receive="Test Page"
        scheduler=rr
        protocol=tcp
        checktype=negotiate

Thanks for the attention,

Fernando.