Syncing a Running Production System to a Running Testing System

Server Configuration

PRODUCTION_SERVER=ticc-1.mines.edu
TEST_SERVER=ticc-2.mines.edu

Login Access

To be able to run an automated sync without a root login it is necessary to create a SSH identity file so that one server can login to the other. The appropriate identity information can be generated by going to the production server and running "sudo ssh-keygen". This command will generate two files: "/root/.ssh/id_rsa" and "/root/.ssh/id_rsa.pub", you must then output the public key to the authorized keys list (in a root environment run "cat /root/.ssh/id_rsa.pub >> /root/.ssh/authorized_keys") and then copy ALL THREE files over to the test server. Important note: normally you would not do this exactly as specified, but since you are syncing these system completely it is important that both systems have all the same identity credentials.

Configuring the Network Interfaces

/etc/udev/rules.d/70-persistent-net.rules :

# This file was automatically generated by the /lib/udev/write_net_rules
# program, run by the persistent-net-generator.rules rules file.
#
# You can modify it, as long as you keep each rule on a single
# line, and change only the value of the NAME= key.

#####################################
## Production Server Network Cards ##
#####################################
# PCI device 0x14e4:0x1678 (tg3)
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="1c:c1:de:ea:64:b8", ATTR{dev_id}=="0x0", ATTR{type}=="1", KERNEL=="eth*", NAME="eth0"
# PCI device 0x14e4:0x1678 (tg3)
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="1c:c1:de:ea:64:b9", ATTR{dev_id}=="0x0", ATTR{type}=="1", KERNEL=="eth*", NAME="eth1"

#####################################
##  Testing Server Network Cards   ##
#####################################
# PCI device 0x14e4:0x1678 (tg3)
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="1c:c1:de:XX:XX:XX", ATTR{dev_id}=="0x0", ATTR{type}=="1", KERNEL=="eth*", NAME="eth2"
# PCI device 0x14e4:0x1678 (tg3)
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="1c:c1:de:XX:XX:XX", ATTR{dev_id}=="0x0", ATTR{type}=="1", KERNEL=="eth*", NAME="eth3"

/etc/network/interfaces :

# This file describes the network interfaces available on your system
# and how to activate them. For more information, see interfaces(5).

# The loopback network interface
auto lo
iface lo inet loopback

# The production server's IP configuration
auto eth0
iface eth0 inet static
address 138.67.36.184
netmask 255.255.192.0
gateway 138.67.1.1

# The test server's IP configuration
auto eth3
iface eth3 inet static
address 138.67.XXX.XXX
netmask 255.255.192.0
gateway 138.67.1.1

Copying Permissions

Before performing a full sync it is necessary to copy the permissions files from the production server to the test server. If you do not do this then when you run rsync it will not set the correct permissions for any files created by users that are new since the last sync.

So, on the production server run:

sudo rsync -avz /etc/passwd /etc/group root@${TEST_SERVER}:/etc/

Syncing the System

Create an exclude file that does not copy running system-critical files (/root/.sync-exclude):

/dev
/proc
/sys
/etc/hosts
/etc/hostname
<pre>

Maybe exclude?:
<pre>/root/.ssh/known_hosts

Then on the production server run:

sudo rsync -avz --exclude-from=/root/.sync-exclude / root@${TEST_SERVER}:/

What this does

For a detailed overview of the commands involved with rsync, use man rsync.

The exclude commands are to exclude files that are not data files. They are all files that we felt did not need to be exact between the two servers since they deal either with temporary files or, in the case of the "known_hosts" file, involve files that should not be the same between the two. Otherwise, this should copy exactly the filesystem between the two.