Distributed Replicated Block Device is a great way to replicate filesystem over the network to another server or host. Distributed Replicated Block Device actually mirrors a whole block device over the network therefore it is often used for High Availability (HA) clusters. Distributed Replicated Block Device is actually a network based RAID 1.

Distributed Replicated Block Device has a number of advantages over a conventional shared storage solutions – read more HERE.

REQUIREMENTS:

• additional disk for synchronization on BOTH MACHINES (preferably same size)
• network connectivity between machines
• working DNS resolution (can fix with /etc/hosts file)
• NTP synchronized times on both nodes

Let’s configure DRBD on Ubuntu Server 12.04!

1. Install following package (BOTH NODES)

root@foo1:~# apt-get install python-software-properties

root@foo2:~# apt-get install python-software-properties

2. Add PPA reporitory for DRBD (BOTH NODES)

Since the Distributed Replicated Block Device kernel module shipped with latest Ubuntu Server 12.04 is 8.4.2 version and only available Distributed Replicated Block Device utils from Ubuntu repositories are 8.3 version (did not get it to work) we need to add a Personal Package Archive to get the 8.4.3 version of Distributed Replicated Block Device utils.

root@foo1:~# add-apt-repository ppa:icamargo/drbd

root@foo2:~# add-apt-repository ppa:icamargo/drbd

1. Update repo information (BOTH NODES)

root@foo1:~# apt-get update

root@foo2:~# apt-get update

2. Install required packages (BOTH NODES)

root@foo1:~# apt-get install drbd8-utils

root@foo2:~# apt-get install drbd8-utils

3. Insert module (BOTH NODES)

root@foo1:~# modprobe drbd
root@foo1:~# lsmod |grep drbd
drbd                  273218  0 
lru_cache              14731  1 drbd
libcrc32c              12543  1 drbd

root@foo2:~# modprobe drbd
root@foo2:~# lsmod |grep drbd
drbd                  273218  0 
lru_cache              14731  1 drbd
libcrc32c              12543  1 drbd

4. Create resource configuration file (BOTH NODES)

We need to create Distributed Replicated Block Device resource configuration file where we define the disks we want to replicate and both nodes IP address and hostname. The newly created resource configuration file must be located at /etc/drbd.d/ directory and have .res file extension. Resource configuration files must be the same on both Distributed Replicated Block Device nodes – we will copy it in our next step! Our resource file /etc/drbd.d/disk1.res is as follows:

resource disk1
{
   startup {
   wfc-timeout 30;
   outdated-wfc-timeout 20;
   degr-wfc-timeout 30;
 }

net {
   cram-hmac-alg sha1;
   shared-secret sync_disk;
 }

syncer {
   rate 100M;
   verify-alg sha1;
 }

on foo1.geekpeek.net {               # Node1 defined
   device /dev/drbd0;
   disk /dev/sdb;                    # Device to use with DRBD
   address 192.168.1.100:7789;       # IP Address and port of Node1
   meta-disk internal;
 }

on foo2.geekpeek.net {               # Node2 defined
   device /dev/drbd0;
   disk /dev/sdb;                    # Device to use with DRBD
   address 192.168.1.101:7789;       # IP Address and port of Node2
   meta-disk internal;
 }
}

5. Edit /etc/hosts file (BOTH NODES)

If you do not have a working DNS resolution edit your hosts file and enter the IP address and FQDN of both nodes!

192.168.1.100 foo1 foo1.geekpeek.net
192.168.1.101 foo2 foo2.geekpeek.net

6. Configure NTP synchronization (BOTH NODES)

Put this in your /etc/crontab file:

1 * * * * root ntpdate your.ntp.server

7. Initialize meta data (BOTH NODES)

root@foo1:~# drbdadm create-md disk1

root@foo2:~# drbdadm create-md disk1

8. Start services (BOTH NODES)

root@foo1:~# /etc/init.d/drbd start

root@foo2:~# /etc/init.d/drbd start

You might get a warning that versions of module and userland are not matching – as long as Distributed Replicated Block Device resource is up and connected it is OK.

9. Check resource

Check if DRBD resource “disk1” is up and connected.

root@foo1:~# cat /proc/drbd
 version: 8.4.2 (api:1/proto:86-101)
 srcversion: 18C7EBE1B3F8CCCB5CF512C
 0: cs:Connected ro:Secondary/Secondary ds:Inconsistent/Inconsistent C r-----
 ns:0 nr:0 dw:0 dr:0 al:0 bm:0 lo:0 pe:0 ua:0 ap:0 ep:1 wo:f oos:1048508

10. Make one node primary

We need to make one Distributed Replicated Block Device node primary and wait for synchronization.

root@foo1:~# drbdadm -- --overwrite-data-of-peer primary all

root@foo1:~# cat /proc/drbd 
version: 8.4.2 (api:1/proto:86-101)
srcversion: 18C7EBE1B3F8CCCB5CF512C 
 0: cs:SyncSource ro:Primary/Secondary ds:UpToDate/Inconsistent C r---n-
    ns:262000 nr:0 dw:0 dr:269976 al:0 bm:15 lo:1 pe:2 ua:8 ap:0 ep:1 wo:f oos:788412
    [====>...............] sync'ed: 25.0% (788412/1048508)K
    finish: 0:00:09 speed: 86,696 (86,696) K/sec

root@foo1:~# cat /proc/drbd 
version: 8.4.2 (api:1/proto:86-101)
srcversion: 18C7EBE1B3F8CCCB5CF512C 
 0: cs:Connected ro:Primary/Secondary ds:UpToDate/UpToDate C r-----
    ns:1048508 nr:0 dw:0 dr:1049172 al:0 bm:64 lo:0 pe:0 ua:0 ap:0 ep:1 wo:f oos:0

11. Make filesystem on primary node

root@foo1:~# mkfs.ext4 /dev/drbd0

12. Mount disk on primary node

root@foo1:~# mount /dev/drbd0 /mnt/
root@foo1:~# df -h
Filesystem      Size  Used Avail Use% Mounted on
/dev/sda1       7.3G 1008M  5.9G  15% /
udev            240M  4.0K  240M   1% /dev
tmpfs           100M  288K   99M   1% /run
none            5.0M  8.0K  5.0M   1% /run/lock
none            248M     0  248M   0% /run/shm
/dev/drbd0      992M  1.3M  940M   1% /mnt

We are now replicating /dev/drbd0 to our second machine foo2.geekpeek.net. In order to mount /dev/drbd0 on the second node we must switch primary -> secondary and secondary -> primary and then mount device.