Please contact your support team if you have a question or need assistance for any Rackspace products, services, or articles.
I recently added a new 50 GB volume to my rackspace server. I previously had one volume and now have 2 volumes. I have a few questions around this addition
Looking forward to some amazing answers!
Thank you for writing in to seek clarification. I will answer your question in turn, with what can be done and what you should do from a best practice perspective.
1. The answer to this question is no it will not. Adding an additional volume adds a new unique block of space to your server. It is up to you to add it within the O.S., format it, assign a drive letter, and determine how best to use this new space.
2. Yes, this is possible by way of using a spanned volume, striped volume, or striped volume with parity.
3. Yes you would need to use a RAID configuration. This is not pre-installed but is configurable within the operating system.
4. Correct, RAID 0 does not provide any redundancy. My recommendation would be to go with a RAID 5 rather than a RAID 10 due to the fac tthat you would need some 3rd party software to configure a software based RAID 10 and a RAID 5 only requires a minimum of 3 disks. Setup of a RAID 5 will be entirely dependent on which operating system you are using. You can use your favorite search engine to find instructions for configuring a RAID 5 on your particular operating system.
- Spanned.Striped Volume - These types of volumes essentially treat all spanned/striped volumes as a single volume. If you have an existing 50 GB volume and you span that with a new 50 GB volume, you will be left with 100 GB of space. The operating system will write to the span as if it is a single disk. The downside to this configuration is that if one volume becomes corrupt or damaged, all data on that volume can be lost. There are some technical differences between spanned and striped volumes but I will leave them out of this discussion.
- Striped Volume With Parity - This volume is discussed separately because this is the recommended solution which provides data redundancy. This is what is known as a RAID 5. This configuration requires at least three volumes but you only get two volumes of usable space. This is due to the fact that as data is written, parity information is also written so that if one disk fails, the other disks along with the parity information can retain the data. This also allows a failed disk to be rebuilt from the parity information. This obviously costs more because you need an extra disk worth of space to store all parity information but it protects against data loss.
I am assuming your line of questioning is centered around a cloud server. In this case spanning, striping, and striping with parity are configured within the operating system. This is not something that is configured dynamically just by way of adding a volume to a server.
I hope this provides the information you are looking for.
Please let us know if we can be of additional assistance.
First of all thank you for your reply.
I am interested in the 'Span' solution for the immediate future. Can you tell me how to make my two volumes behave like a single volume from the perspective of the OS.
For the long term i will be interested in a RAID5/ RAID10 solution. Please let me know how to create this configuration as well on my OS?
Spanning and striping WITHOUT parity is not recommended due to the potential for data loss. If you are insistent on spanning, the process is similar to the one described here -> https://support.rackspace.com/how-to/using-automatic-and-manual-disk-partitioning-on-cloud-servers/
If you are using Windows, Microsoft discusses this here -> https://technet.microsoft.com/en-us/library/cc772180%28v=ws.11%29.aspx
If we can be of further assistance please let us know.
I also wanted to know that i set up my new volume and did mkfs on the new volume as mentioned here -> https://support.rackspace.com/how-to/prepare-your-cloud-block-storage-volume/
So now will data start getting saved on this new disk or do i have to create a volume group etc. as mentioned here -> https://www.cyberciti.biz/faq/howto-add-disk-to-lvm-volume-on-linux-to-increase-size-of-pool/
My area of expertise is Windows so I cannot speak to the process necessary to create a spanned volume in Linux. While the steps in the resource referenced appear to be in line with my understanding of the process in Linux, I cannot say they are 100% accurate as that is outside my area of expertise. I would advise that you make sure you have a full backup of any data that may be lost should you run in to an issue while trying to create a new volume group and logical volume.
One more question.
Since my current disk was not using LVM is it better to just ask Rackspace for a 100GB(since on the console i can only get 50/75GB disks) drive and clone my current volume into the new volume.
Generally speaking, the best approach to use is to not use the system drive for data but instead use a CBS volume for your data. This way, if you ever have a problem with the server and have to spin up a new one you can detach the CBS volume from one server and attach it to another without the need to restore data from a backup or migrate the data in some other fashion. From previous discussions, another option for you to use is a cloud database. Cloud databases provide the elasticity it appears you are trying to solve for as the space can be dynamically adjusted as your needs for additional storage grow.
I hope this provides the information you were looking for.