XenServer 5.6 FP1 add local drive to backup VMs with OpenFiler
The goal of the following blog is to have a cheap backup system for my XenServer @ Home. Remember snapshots are not backup and it’s not recommended to backup you VMs the way I will describe it in a production environment.
You can find a Citrix KB to add local storage here : http://support.citrix.com/article/CTX121313
The idea is to install a new hard drive in my XenServer @ Home, configure it and host an OpenFiler VM to share a NFS volume. As the NFS volume will be shared, I want to use it as a backup share to archive all my VMs snapshots with the VM Protection Policies option in Citrix XenServer 5.6 FP1. This option (VM Protection Policies) is available only with the platinum XenServer license.
So I plugged the new hard drive and plug it in my XenServer @ Home and switch on the box, then to enable and activate the drive, I needed first to check XenServer was able to see the new hard drive, i used the fdisk -l command :
[root@suomixen ~]# fdisk -l
Disk /dev/sda: 1000.2 GB, 1000204886016 bytes
255 heads, 63 sectors/track, 121601 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Device Boot Start End Blocks Id System
/dev/sda1 * 1 523 4194304 83 Linux
Partition 1 does not end on cylinder boundary.
/dev/sda2 523 1045 4194304 83 Linux
/dev/sda3 1045 121601 968371393 8e Linux LVM
Disk /dev/sdb: 1000.2 GB, 1000204886016 bytes
255 heads, 63 sectors/track, 121601 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk /dev/sdb doesn't contain a valid partition table
My new hard drive is here, I can notice the /dev/sdb which is empty. Then I needed to create a physical volume by using the pvcreate /dev/sdb command line :
[root@suomixen ~]# pvcreate /dev/sdb
Physical volume "/dev/sdb" successfully created
Now the physical volume created I can create a logical volume by using the xe sr-create command line. Before I need to identify the ID of the new volume created with the xe pbd-list command :
[root@suomixen ~]# xe pbd-list
uuid ( RO) : ea29b92e-ad26-9d80-92e1-b150d689b3aa
host-uuid ( RO): cd260b4c-f311-4fae-852b-1b794bc8e276
sr-uuid ( RO): 862abbc4-b86e-bc98-c784-e24e323602c0
device-config (MRO): device: /dev/disk/by-id/scsi-SATA_ST31000340NS_9QJ2EXS0
currently-attached ( RO): true
uuid ( RO) : b990f978-d89c-fa6a-b291-3d1f15245ce3
host-uuid ( RO): cd260b4c-f311-4fae-852b-1b794bc8e276
sr-uuid ( RO): 65e04d65-f30e-401d-18e3-ccf479a50bf4
device-config (MRO): location: /opt/xensource/packages/iso; legacy_mode: true
currently-attached ( RO): true
uuid ( RO) : 7397b1ec-d2e2-3681-51ef-a213806d9fc5
host-uuid ( RO): cd260b4c-f311-4fae-852b-1b794bc8e276
sr-uuid ( RO): fc5a979a-f6a4-7352-afab-296c149e8907
device-config (MRO): serverpath: /mnt/backup_group/backup_vms/Backup_Xen/; server: 192.168.0.98; options:
currently-attached ( RO): true
uuid ( RO) : ea85c804-5898-e449-342e-96f5c3dd9b44
host-uuid ( RO): cd260b4c-f311-4fae-852b-1b794bc8e276
sr-uuid ( RO): 829d7f51-0bc8-4ee9-66a4-5c8b8fc17423
device-config (MRO): username: suomi\administrator; cifspassword_secret: 4e912a17-a7c8-fb85-dff1-491460cdbf2c; SRmaster: true; type: cifs; location: //192.168.0.100/Isos
currently-attached ( RO): true
uuid ( RO) : 6aaa8610-ccc6-0ba9-0ed9-747128b5bd27
host-uuid ( RO): cd260b4c-f311-4fae-852b-1b794bc8e276
sr-uuid ( RO): 2b210a05-a1ff-01cd-62e7-8bf864f5e1b1
device-config (MRO): location: /dev/xapi/cd
currently-attached ( RO): true
uuid ( RO) : eed7d2dc-f7da-7149-28da-26a02d9f5fe7
host-uuid ( RO): cd260b4c-f311-4fae-852b-1b794bc8e276
sr-uuid ( RO): 79c6b13b-3876-64b1-e312-e97a87a9ab9e
device-config (MRO): device: /dev/disk/by-id/scsi-SATA_ST31000340NS_9QJ1CRYN-part3
currently-attached ( RO): true
uuid ( RO) : 5c190639-7111-6fe3-b38d-262e608b64e1
host-uuid ( RO): cd260b4c-f311-4fae-852b-1b794bc8e276
sr-uuid ( RO): 963e166b-763a-2b33-1117-e131f732542b
device-config (MRO): location: /dev/xapi/block
currently-attached ( RO): true
I can see my new volume, the path and its name, /dev/disk/by-id/scsi-SATA_ST31000340NS_9QJ2EXS0, the first one in the previous list. Now I can create the logical volume to make it ready to execute my backup plan idea, I use the xe sr-create command line.
[root@suomixen disk]# xe sr-create type=lvm content-type=user device-config:device=/dev/disk/by-id/scsi-SATA_ST31000340NS_9QJ2EXS0 name-label=LOCAL
862abbc4-b86e-bc98-c784-e24e323602c0
You need to use the command bellow but don’t forget to use your device ID, not mine, it won’t work
xe sr-create type=lvm content-type=user device-config:device=/dev/disk/by-id/YOURDEVICENAME name-label=MY_NEW_VOLUME_NAME and you should be just fine. The installation of the new drive is done and now we can download the OpenFiler appliance for XenServer here : http://www.openfiler.
com/community/download/. Once downloaded, import the VM into XenServer and start it.
To import the Openfiler XenServer appliance you can follow the instruction or you can skip this small part if you know how to proceed.
Uncompress the archive you downloaded and in the Citrix XenCenter console do as follow :
That’s it, now you just need to wait a bit for the file to be send to the XenServer machine. Once the machine is started you will add a drive (in my case, big enough to backup the VMs snapshots 900GB)
Previous steps show you how to add a drive to your OpenFiler XenServer appliance. Don’t forget to reboot the OpenFiler. Now we can configure OpenFiler. To do so, go on the console tab in your XenCenter and check the main screen, you should find the following information :
Now you have the address to connect the we management interface, go on : https://the.ip.address:446
OpenFiler default UserName is openfiler and the default password is password. Proceed to logon and click on the system tab to configure the network regarding your own network settings:
Don’t forget to configure the network access as you can see on the previous screen shot on the bottom of the screen. Next we will configure the volumes and everything need to be setup to enable a NFS share, click on the Volumes tab and follow the next steps :
Now click on the Services tab and enable SMB / CIFS Server and NFS v3 Server :
Click now the the Shares tab to create and configure a share :
Chose a name for your share and click on Create Sub-Folder, the new folder is created so click on it :
Click on Make Share to enable the share of this new folder and continue the setup in the next window, go at the bottom of this page to find the Host Access Configuration part and change the option bellow :
Click on Update
Time to make a check point :
- The new hard drive has been install in the XenServer box
- The new hard drive has been setup and configure in XenServer
- The OpenFiler XenServer Appliance is up and running
- The OpenFiler XenServer Appliance is setup
- A new NFS share has been configured
Everything looks good so far, now the next steps are :
- Adding the NFS Share to XenServer as a New Storage
- Adding and configuring new VM Protection Policies
- Test everything
Adding the new NFS Share as a New Storage in XenServer is not mandatory to use it with VM protection Policies, I just do it to test the access and make sure everything is working well. In XenCenter proceed by following the next screen shots :
That’s it.
Now it’s time to configure the VM Protection Policies option, I do it all through XenCenter (I haven’t check how to script it yet…) to do so, follow the way :
And in the next screen is about the archive schedule backups :
Don’t forget to test the NFS share and click on Next
And Finish. AsI check th bow to run the job when I clicked on finish I’ve been able to see very quickly the snapshot result and everything went fine :
Of course the schedule archive backup on the NFS Share will take much more time so if you want to check if your job is still on the way, you can click on the Pool and chose the Logs tab :
This is copying, the operation is a success, it takes some time to copy the scheduled backup but it works. For example, the VM I wanted to backup for all the previous screen shot took 5 second to snapshot but 1h30 to copy. I had some issue trying to do the same with more than one VM at the same time, for an unknown reason snapshots were fine but the copy stopped. I chose to backup two VMs per night over the week and it works well. Finally, you might have notice you can copy your schedule snapshots only on a CIFS or a NFS share, nothing else for now.
Sources :
OpenFiler Website

Host : suomixen01
Cpu Nbr : 8
Memory : 33Gb
Virtual Machines : 42
Powered On VMs : 20
Halted VMs : 22
Suspended VMs : 0











































7 Comments for "XenServer 5.6 FP1 add local drive to backup VMs with OpenFiler"
Pingback: Adding local storage to XenServer 5.x | ArchY.net
Pingback: 07-03-2011 weekly tweets | ArchY.net
You should try Nexenta appliance in order to have ZFS storage & Deduplucation.
(Fred aka Andreas Kisser)
Hi Fred,
you can select zfs instead of ext3 as I did with OpenFiler. But thanks for the tip, I will definitely check Nexenta.
Cheers !
Hi Stephane,
very nice article. I did something similiar for backing up my VMs.
I had a spare external USB hard drive which I wanted to use for this. As I didn’t want to void the guarantee I wasn’t able to take the drive put it in the server itself.
So I installed a Windows VM, attached the USB drive to the XenServer and mapped the USB drive to the VM. Then I formatted it with NTFS and shared it as CIFS share.
As my XenServer can access this CIFS share I used it as destination for my VM Protection Policies.
The back up process may be a bit slower then your solution but I can access and recover my back ups from every Windows machine as it’s an external NTFS formatted drive.
Cheers,
Tim
Hi Stephane,
good idea. I also did it with a local attached USB drive at the xen server and a free backup script. Taking Snapshot, copying snapshot with smb to usb disk and deleting snapshot. works great.
What if your xenserver and the open filer vms dies. Can you attach the open filer volumen to a new openfiler vm again and access the backuped vms ?
greetings,
Marcus
Pingback: PHD Virtual Backup for XenServer | archy.net
Leave a Comment
If you liked our post and you want to share your thought with us dont hesitate to leave a commnet, the mail will not be published.