[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

[Linux-cluster] RHCS simple ip-failover problem

Dear List,

I've received today a evaluation License for RHEL 6.1[Red Hat Enterprise Linux Server release 6.1 (Santiago)] for learning purposes with the RHCS.

So I've setup 2 machines vbox1 and vbox2, with 2 Interfaces (Intranet + Hearbeat) and created a cluster.conf with a simple IP-fail-over Scenario.
Starting the cluster no node takes the IP.
After checking cluster state I'm a little bit confused, it doesn't list the service.
Maybe i understood something wrong from the documentation.

[root vbox1 cluster]# clustat  -l
Cluster Status for vbox @ Sun Nov 27 14:38:00 2011
Member Status: Quorate

 Member Name                             ID   Status
 ------ ----                             ---- ------
 vbox1.example.local                         1 Online, Local
 vbox2.example.local                         2 Online

By testing the rules the system takes the ip-address.
[root vbox1 cluster]# rg_test test /etc/cluster/cluster.conf start service webprod
Running in test mode.
Starting webprod...
<debug>  Link for eth0: Detected
Link for eth0: Detected
<info>   Adding IPv4 address to eth0
Adding IPv4 address to eth0
<debug>  Pinging addr from dev eth0
Pinging addr from dev eth0
<debug>  Sending gratuitous ARP: 00:0c:29:00:d1:05 brd ff:ff:ff:ff:ff:ff
Sending gratuitous ARP: 00:0c:29:00:d1:05 brd ff:ff:ff:ff:ff:ff
rdisc: no process killed
Start of webprod complete

[root vbox1 cluster]# ip addr list eth0
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 00:0c:29:00:d1:05 brd ff:ff:ff:ff:ff:ff
    inet brd scope global eth0
    inet scope global secondary eth0
    inet6 fe80::20c:29ff:fe00:d105/64 scope link
       valid_lft forever preferred_lft forever

maybe someone give me a point to the right direction?

# Intranet vbox1.example.local vbox1 vbox2.example.local vbox2
# Heartbeat h-vbox1.example.local h-vbox1 h-vbox2.example.local h-vbox2
# Service-IP vbox.example.local vbox

/etc/cluster/cluster.conf (for testing purposes only manual fencing. later i'l try it with fence_ipmilan)

<?xml version="1.0"?>
<cluster config_version="3" name="vbox">
<cman expected_votes="1" two_node="1"/>
<clusternode name="vbox1.example.local" nodeid="1">
<altname name="h-vbox1.example.local"/>
<method name="n1">
<device name="human" nodename="vbox1.example.local"/>
<clusternode name="vbox2.example.local" nodeid="2">
<altname name="h-vbox2.example.local"/>
<method name="n2">
<device name="human" nodename="vbox2.example.local"/>
<fencedevice agent="fence_manual" name="human"/>
<ip address="" monitor_link="on" sleeptime="15"/>
<service autostart="1" domain="web" exclusive="0" name="webprod" recovery="restart">
<ip ref=""/>
<failoverdomain name="web" nofailback="0" ordered="1" restricted="0">
<failoverdomainnode name="vbox1.example.local" priority="1"/>
<failoverdomainnode name="vbox2.example.local" priority="2"/>

Thanks in advance

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]