Active Passive setup with heartbeat drbd and Apache

Heartbeat 2 + DRBD 8 + Apache

Step 1 : Configure DRBD -> Follow DRBD configuration steps

Step 2 : Give Virtual IP for HeartBeat

###At both machines

ifconfig eth0:1 192.168.91.95 netmask 255.255.255.0 (put at /etc/rc.d/rc.local)

Step 3 : Configure Apache

Edit /etc/httpd/conf/httpd.conf

Machine – 192.168.91.2

<VirtualHost 192.168.91.95>

DocumentRoot /web <-- DRBD

<Directory "/web">

allow from all

Options +Indexes

</Directory>

ServerName 192.168.91.2

</VirtualHost>

Machine – 192.168.91.3

<VirtualHost 192.168.91.95>

DocumentRoot /web <-- DRBD

<Directory "/web">

allow from all

Options +Indexes

</Directory>

ServerName 192.168.91.3

</VirtualHost>

Step 4: Installation and Configuration Heatbeat

###At Both machine

rpm -ivh lm_sensors-2.8.7-2.40.3.i386.rpm

rpm -ivh heartbeat-pils-2.0.8-2.el4.centos.i386.rpm

rpm -ivh heartbeat-stonith-2.0.8-2.el4.centos.i386.rpm

rpm -ivh heartbeat-2.0.8-2.el4.centos.i386.rpm

###Create 3 configuration files at /etc/ha.d/

All three files identical at both machines

File “authkeys”

auth 2

2 crc

Permission 600 for authkeys

File “haresources”

box1.lan drbddisk::r0 Filesystem::/dev/drbd0::/web::ext3 192.168.91.95 httpd

File “ha.cf”

bcast eth0

keepalive 2

warntime 10

deadtime 30

initdead 120

udpport 694

auto_failback off

ping 192.168.91.1

node box1.lan

node box2.lan

use_logd yes

logfile /var/log/hb.log

debugfile /var/log/heartbeat-debug.log

Start Heartbeat

Start Master first

# service heartbeat start

Start DRBD

At master

#modprobe drbd

#drbdadm adjust r0

#drbdsetup /dev/drbd0 primary --overwrite-data-of-peer r0

#mount -o rw /dev/drbd0 /web/

check

#cat /proc/drbd

It should see like

At Standby machine

# modprobe drbd

# drbdadm adjust r0

Check

#cat /proc/drbd

It should see like

Start Apache Server at Both Machines

# service httpd start

Try to simulate stop Master, Slave should take over the Apache.

###Troubleshooting

If “Split-Brain detected, dropping connection! ” error occur

Run drbdadm -- --discard-my-data connect whatever at Slave

You should see log like this.

Sep 17 16:48:01 box2 Filesystem[4768]: [4797]: INFO: Running start for /dev/drbd0 on /web

Sep 17 16:48:01 box2 kernel: kjournald starting. Commit interval 5 seconds

Sep 17 16:48:01 box2 kernel: EXT3 FS on drbd0, internal journal

Sep 17 16:48:01 box2 kernel: EXT3-fs: recovery complete.

Sep 17 16:48:01 box2 kernel: EXT3-fs: mounted filesystem with ordered data mode.

Sep 17 16:48:02 box2 Filesystem[4757]: [4813]: INFO: Success

Sep 17 16:48:03 box2 IPaddr[4826]: [4857]: INFO: Running OK

Sep 17 16:48:04 box2 mach_down[4620]: [4869]: info: /usr/share/heartbeat/mach_down: nice_failback: foreign resources acquired

Sep 17 16:48:04 box2 mach_down[4620]: [4873]: info: mach_down takeover complete for node box1.lan.

Sep 17 16:48:04 box2 heartbeat: [4165]: info: mach_down takeover complete.

###Make sure auto startup

chkconfig drbd on

chkconfig heartbeat on

chkconfig httpd off