Ultramonkey Ping problem

[Lars Roland]

Hi all

I have just configured a Debian SpamAssassin cluster according to the
"Streamline High Availability and Load Balancing" topology
(http://www.ultramonkey.org/3/topologies/sl-ha-lb-eg.html), but I am
having some IP/VIP problems. My IP setup is:

----
VIP 10.10.20.10
Director_1/real_1: 10.10.20.11
Director_2/real_2: 10.10.20.12
----

The problem is that I can ping director_1 from director_2 but not the
other way aound, thus since I use ping as check method, director 1
shows director_2 as down i.e.

----
TCP  10.10.20.10:783 wlc
  -> 10.10.20.11:783              Local   1      0          460
  -> 10.10.20.12:783              Route   0      0          0
----

If I however run the following on director_1

----
ipvsadm -e -t 10.10.20.10:783 -r 10.10.20.12 -w 1
----

then ipvsadm shows

----
TCP  10.10.20.10:783 wlc
  -> 10.10.20.11:783              Local   1      0          460
  -> 10.10.20.12:783              Route   1      0          24
----

and it begins routing traffic successfully (i.e. spamassassin on
10.10.20.12 (director_2) receives connections). I can however still
not contact director_2 directly from director_1 i.e.

----
telnet 10.10.20.12 783
----

gives no reply, I am running kernel 2.6.18 and I have configured
sysctl according to the article in order to fix the Linux arp problem.
I hope some one has an idea on what is wrong, because I am in the
blind here.



--
Lars Roland




PS: My config files are as follows:

/etc/network/interfaces
----
# The loopback network interface
auto lo
iface lo inet loopback
# The primary network interface
auto eth0
iface eth0 inet static
        address 10.10.20.11
        netmask 255.255.0.0
        gateway 10.10.25.254
        dns-nameservers 10.10.25.254
# loopback interface of virtal ip
auto lo:0
iface lo:0 inet static
       address 10.10.20.10
       netmask 255.255.255.255
       up sysctl -p > /dev/null
----


/etc/ha.d/ha.cf
----
# uncoment for extensive degugging
# debug 100
# Facility to use for syslog()/logger
logfacility     local0
# keepalive: how long between heartbeats?
keepalive 2
# deadtime: how long-to-declare-host-dead?
deadtime 10
# warntime: how long before issuing "late heartbeat" warning?
# See the FAQ for how to use warntime to tune deadtime.
warntime 10
# Very first dead time (initdead)
initdead 10
# auto_failback:  determines whether a resource will
# fail back to its "primary" node, or remain on whatever
# node is serving it until that node fails.
auto_failback on
# Set up a broardcast heartbeat medium
bcast eth0
# Tell what machines are in the cluster
node node1.somedomain
node node2.somedomain
----


/etc/ha.d/haresources
----
node1.somedomain 10.10.20.10/24/eth0 ldirectord::ldirectord.cf
----


/etc/ha.d/ldirectord.cf
----
logfile="/var/log/ldirectord.log"
checktimeout=10
checkinterval=2
autoreload=yes
quiescent=yes
virtual = 10.10.20.10:783
        real = 10.10.20.11:783 gate
        real = 10.10.20.12:783 gate
        service =  none
        scheduler = wlc
        protocol = tcp
        checktype = ping
----


-- 
Ultra Monkey - http://www.ultramonkey.org/
To UNSUBSCRIBE, email to lisa at ultramonkey.org, with a body:
unsubscribe ultramonkey-users your-email-address at some.domain
where "your-email-address at some.domain" is YOUR email address.