Openfiler 2.3 Active/Passive Cluster (Heartbeat, DRBD) With Offsite Replication Node
Introduction
Openfiler is a Linux based NAS/SAN application which can deliver storage over nfs/smb/iscsi and ftp. It has a web interface over that you can control these services. The howto is based on theHowto
from Kyle Gililland. A lot of thanks to him for this.
The cluster we build will consist of two nodes replicating each other and taking over services and storage in case of emergency. Furthermore we have an Offsite Replication Server, which ideally stands in a physically different position and replicates the
configurations/storage from which ever node is active. In case of emergency this Offsite Replication Server can be used to restore the cluster and to deliver the services.
I used the Openfiler VMware Images 2.3 x86 and Openfiler 2.3 x86 Installation Media to test out it this howto. Hopefully you will be able to repeat my steps and have a safe and cheap Storage Solution.
Overview
1. Installation
2. Prepaire the partitions for DRBD
3. Create basic heartbeat configuration
4. DRBD Configuration
5. Prepaire the configuration partition
6. Initiate Data/Storage partition
7. Create final heartbeat configuration
8. Enable/Disable sytem services
9. Test failover on node1 and node2
10. Test recover of filer01 and filer02
10.1 DRBD Configuration
10.2 filer01 and filer02 redo configuration
10.3 Retake resources and run cluster again
11. Use Replication Node as Main Node
11.1 Finished Replication, how to turn Replication Node in standby again
12. Add another Storage Partition
Requirements
Hardware
3x boxes that meet the minimum hardware requirements of openfiler
2x ethernet interfaces at each box
1. Installation
In the Installation process of openfiler leave the portion of the disk you wanna use for Storage Services unused. In this case it was a second harddisk in each system. We create the partitions for Storage Replication after the installation.
Networkconfiguration
on filer01:
eth0: 10.10.11.101
eth1: 10.10.50.101
on filer02:
eth0: 10.10.11.102
eth1: 10.10.50.102
on filer03:
eth0: 10.10.11.103
eth1: 10.10.50.103
virtual devices: ( this will be created by heartbeat, dont add these ip's to any interface )
eth0:0 10.10.11.100 ( iscsi target server ip )
eth1:0 10.10.50.100 ( replication interface )
After finishing the installation build yourself a nice working environment while you update the openfiler systems. It's important to update the systems now as openfiler 2.3 standard installation media dont bring DRBD 8.3.x which is needed for 3-node replication.
conary update conary
conary updateall --replace-files --no-conflict-check
While the systems update you add the other filers to the
/etc/hosts file and generate some ssh-keys and exchange them.
root@filer01 ~# nano /etc/hosts
#Installed by rBuilder
127.0.0.1 filer03 localhost.localdomain localhost
10.10.50.101filer01
10.10.50.102filer02
Then we create ssh-keys without a passphrase which we use for easier file-exchange.
root@filer01 ~# ssh-keygen -t dsa
Generating public/private dsa key pair.
Enter file in which to save the key (/root/.ssh/id_dsa):
Created directory '/root/.ssh'.
Enter passphrase (empty for no passphrase):
Enter same passphrase again:
Your identification has been saved in /root/.ssh/id_dsa.
Your public key has been saved in /root/.ssh/id_dsa.pub.
The key fingerprint is:
4c:76:ec:61:10:2a:07:9f:35:8d:9b:16:89:5c:bc:b9 root@filer01
Do the same on the other 2 nodes:
root@filer02 ~# ssh-keygen -t dsa
root@filer03 ~# ssh-keygen -t dsa
Exchange the public keys between the nodes:
root@filer01 ~# scp .ssh/id_dsa.pub root@filer02:~/.ssh/authorized_keys2
2. Prepare The Partitions For DRBD
Setup the partitions for DRBD which will contain the configuration files for openfiler and its storage services.Don't add any of this to /etc/fstab. If you created the partitions in the installation process remove the entries
for them at the/etc/fstab file.
You have todo this steps on all 3 Machines.
fdisk /dev/sdb
Command (m for help): n
Command action
e extended
p primary partition (1-4)
p
Partition number (1-4): 1
First cylinder (1-1566, default 1):
Using default value 1
Last cylinder, +cylinders or +size{K,M,G} (1-1566, default 1566): 67
Command (m for help): n
Command action
e extended
p primary partition (1-4)
p
Partition number (1-4): 2
First cylinder (68-1566, default 68):
Using default value 68
Last cylinder, +cylinders or +size{K,M,G} (68-1566, default 1566):
Using default value 1566
Command (m for help): t
Partition number (1-4): 2
Hex code (type L to list codes): 8e
Changed system type of partition 2 to 8e (Linux LVM)
You have to exist fdisk with w to write the changed to the disk.
You can check the partitions after this:
fdisk -l
Disk /dev/sda: 2355 MB, 2355978240 bytes
16 heads, 63 sectors/track, 4565 cylinders
Units = cylinders of 1008 * 512 = 516096 bytes
Disk identifier: 0x00000000
Device Boot Start End Blocks Id System
/dev/sda1 * 1 4565 2300696 83 Linux
Disk /dev/sdb: 12.9 GB, 12884901888 bytes
255 heads, 63 sectors/track, 1566 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0xff9e0345
Device Boot Start End Blocks Id System
/dev/sdb1 1 67 538146 83 Linux
/dev/sdb2 68 1566 12040717+ 8e Linux LVM
NOTE: Before you can continue the systems should have finished update.
3. Create A Basic Heartbeat Configuration
This step is needed to create the virtual IP that DRBD is going to use for the offsite replication.
Create /etc/ha.d/authkeys ( filer01 andfiler02 ):
auth 2
2 crc
Change the accessrights to the /etc/ha.d/authkeys file:
chmod 600 /etc/ha.d/authkeys
Create etc/ha.d/ha.cf ( filer01 andfiler02 ):
<?xml version="1.0" ?>
<cluster>
<clustering state="on" />
<nodename value="filer01" />
<resource value="MailTo::it@company.com::ClusterFailover"/>
<resource value="IPaddr::10.10.50.100/24/eth1" />
</cluster>
Go to https://10.10.11.101:446 and start iscsi target service ( this will create/etc/ha.d/haresources on
filer01 ).
Copy haresource file to filer02 and start heartbeat IP:
root@filer01 ~# scp /etc/ha.d/haresources root@filer02:/etc/ha.d/haresources
root@filer01 ~# service heartbeat start
root@filer02 ~# service heartbeat start
Wait till heartbeat is up on both nodes, then stop heartbeat on
filer02:
root@filer02 ~# service heartbeat stop
This should create a failover where filer01 gets the ip address10.11.50.100, check this with
ifconfig onfiler01, the output should be like this.
root@filer01 ~# ifconfig
eth0 Link encap:Ethernet HWaddr 00:0C:29:D9:66:27
eth1:0 Link encap:Ethernet HWaddr 00:0C:29:D9:66:31
inet addr:10.10.50.100 Bcast:10.10.50.255 Mask:255.255.255.0
If this is set you can now continue to start creating the drbd devices. If not check the output of/var/log/ha-log and try to figure out what is wrong.
4. DRBD Configuration
This manual implies that you have worked with drbd in the past and know howto tune the drbd common section regarding to your hardware. Using protocol C is save on stacked resources but not the best regarding network performance.
Remove the old drbd.conf and create a new one:
root@filer01 ~# rm /etc/drbd.conf
root@filer01 ~# nano /etc/drbd.conf
global {
usage-count yes;
}
common {
protocol C;
disk {
on-io-error detach;
}
net {
after-sb-0pri disconnect;
after-sb-1pri disconnect;
after-sb-2pri disconnect;
rr-conflict disconnect;
}
syncer {
al-extents 257;
rate 100M;
}
handlers {
pri-on-incon-degr "echo O > /proc/sysrq-trigger ; halt -f";
pri-lost-after-sb "echo O > /proc/sysrq-trigger ; halt -f";
local-io-error "echo O > /proc/sysrq-trigger ; halt -f";
}
startup {
degr-wfc-timeout 120; # 2 minutes.
}
}
resource meta {
on filer01 {
device /dev/drbd0;
disk /dev/sdb1;
address 10.10.50.101:7788;
meta-disk internal;
}
on filer02 {
device /dev/drbd0;
disk /dev/sdb1;
address 10.10.50.102:7788;
meta-disk internal;
}
}
resource data {
on filer01 {
device /dev/drbd1;
disk /dev/sdb2;
address 10.10.50.101:7789;
meta-disk internal;
}
on filer02 {
device /dev/drbd1;
disk /dev/sdb2;
address 10.10.50.102:7789;
meta-disk internal;
}
}
resource meta-U {
stacked-on-top-of meta {
device /dev/drbd10;
address 10.10.50.100:7788;
}
on filer03 {
device /dev/drbd10;
disk /dev/sdb1;
address 10.10.50.103:7788;
meta-disk internal;
}
}
resource data-U {
stacked-on-top-of data {
device /dev/drbd11;
address 10.10.50.100:7789;
}
on filer03 {
device /dev/drbd11;
disk /dev/sdb2;
address 10.10.50.103:7789;
meta-disk internal;
}
}
Copy this file over to the other nodes:
root@filer01 ~# scp /etc/drbd.conf root@filer02:/etc/drbd.conf
root@filer01 ~# scp /etc/drbd.conf root@filer03:/etc/drbd.conf
Initiate the lower resources:
root@filer01 ~# drbdadm create-md meta
root@filer01 ~# drbdadm create-md data
root@filer02 ~# drbdadm create-md meta
root@filer02 ~# drbdadm create-md data
Start DRBD on filer01 and filer02:
root@filer01 ~# service drbd start
root@filer02 ~# service drbd start
Set the lower drbd resources primary on filer01:
root@filer01 ~# drbdsetup /dev/drbd0 primary -o
root@filer01 ~# drbdsetup /dev/drbd1 primary -o
Create the DRBD Metadata on the stacked ressource:
root@filer01 ~# drbdadm --stacked create-md meta-U
root@filer01 ~# drbdadm --stacked create-md data-U
Enable the stacked resource and set it to primary:
root@filer01 ~# drbdadm --stacked up meta-U
root@filer01 ~# drbdadm --stacked up data-U
root@filer01 ~# drbdsetup /dev/drbd10 primary -o
root@filer01 ~# drbdsetup /dev/drbd11 primary -o
Create DRBD metadata on filer03 and start the resources:
root@filer03 ~# drbdadm create-md meta-U
root@filer03 ~# drbdadm create-md data-U
root@filer03 ~# service drbd start
Control the sync process on the nodes with:
root@filer01 ~# service drbd status
drbd driver loaded OK; device status:
version: 8.3.7 (api:88/proto:86-91)
GIT-hash: ea9e28dbff98e331a62bcbcc63a6135808fe2917 build by phil@fat-tyre, 2010- 01-13 17:17:27
m:res cs ro ds p mounted fstyp e
0:meta Connected Primary/Secondary UpToDate/UpToDate C
1:data Connected Primary/Secondary UpToDate/UpToDate C
10:meta-U^^0 Connected Primary/Secondary UpToDate/UpToDate C
11:data-U^^1 Connected Primary/Secondary UpToDate/UpToDate C
This way it should look when all nodes are synced and filer01 is the primary.
5. Prepare The Configuration Partition
Filer01
Initiate meta partition:
NOTE: Here we gonna use the stacked resource instead of the lower level one!
root@filer01 ~# mkfs.ext3 /dev/drbd10
Openfiler to Meta-Partition:
root@filer01 ~# mkdir /meta
root@filer01 ~# ln -s /meta/etc/proftpd/ /etc/proftpd
httpd Modules for Openfiler ( use lib64 for 64bit openfiler ):
root@filer01 ~# rm /opt/openfiler/etc/httpd/modules
root@filer01 ~# ln -s /usr/lib/httpd/modules /opt/openfiler/etc/httpd/modules
Restart Openfiler to see if everything worked:
root@filer01 ~# service openfiler restart
NOTE: If it doesnt restart check if you have linked the httpd Modules, probably you use lib64 instead of lib!
Filer02 and Filer03
Openfiler Configuration:
mkdir /meta
mv /opt/openfiler/ /opt/openfiler.local
ln -s /meta/opt/openfiler /opt/openfiler
Samba/NFS/ISCSI/PROFTPD Configuration Files to Meta Partition:
service nfslock stop
QQ群⑧：
窥视卡
雷达卡
发表于 2015-11-21 10:23:40
提升卡
置顶卡
沉默卡
喧嚣卡
变色卡
千斤顶
显身卡