Server Fault is a question and answer site for system and network administrators. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I just started the journey into cloud and my choice was Amazon. The free tier was very cool so i opted for this stuff.

In the free tier there was a free storage of 30Gb (EBS), so i made a EBS Volume of 15 Gb for a start and attached it to an instance i created (Ubuntu 12.04).

When i try to find the disk under /dev/sda(xvda) or /dev/sdf(xvdf)

There is no disk mounted at all. I also tried mount -a but no use on that either. I did not actually create any 8 Gb disk but yet there is a 8 gb disk with the OS.

Before i move my sites to aws and burn my hands, i wanted to know if the disk of 8Gb can be expanded or not. If in future the 8 Gb gets full then it would be difficult moving the sites to something other provider.

Apart from creating snapshots and mounting them on a bigger disk is there any process to increase the disk space without effecting anything on the disk (Websites, databases etc).

Please clarify the EBS terminologies and processes.

share|improve this question

Yes, you can change the EBS root volume to a size larger than 8 GB, but it requires a bit of work including stopping your instance for a few minutes. The basic procedures is:

  1. stop instance
  2. detach EBS volume
  3. snapshot EBS volume
  4. create new, larger EBS volume from snapshot
  5. attach new EBS volume
  6. start instance
  7. resize file system

I have written an article describing this process in more detail including sample commands:

That said, I generally recommend storing your data on an EBS volume separate from the root EBS volume. You can use similar processes to resize the data volume.

share|improve this answer
Thanks Eric, the guide is very informative. I will try that on my ec2 right away. – Sameer Manas May 9 '12 at 18:57
Sameer asked "Apart from creating snapshots and mounting them on a bigger disk is there any process to increase the disk space without effecting anything on the disk (Websites, databases etc)." Your answer, Eric (and BTW, thank you for all the incredible work you've done to make AWS usable), is to create a snapshot and restore it to a bigger disk. You have to do it that way because, no, you cannot increase the size of an EBS volume. Which Sameer says is great news. This is too confusing. – Old Pro May 11 '12 at 2:18

I think you're confusing the ephemeral storage vs the EBS. When you spawn a new isntance in the ec2... your "os" volume is typically put on a "ephemeral storage" which is considered "temporary". I.e. if you switch off your instance... everything in ephemeral is reset to the base image you initially selected. (rebooting isn't the same as "switching off") You can setup EBS backed instances, where the OS is not on ephemeral storage... but rather an EBS volume. In that case, you can switch it off and on at will.. and as long as you don't delete your EBS volume... it remains intact.

As far as expanding volumes goes... I typically just assign a new volume to the instance and use LVM to span the volumes to make one larger volume. You can scale up on the fly as you need... and I have not noticed any significant performance impact. The only down-side is that you really can't scale down.

share|improve this answer
No. If Sameer is trying to fit in the AWS free tier, then he is using a t1.micro EC2 instance. This requires him to run an EBS boot instance which means that he is not running the OS on an ephemeral (instance-store) disk. – Eric Hammond May 9 '12 at 16:54
Yes Eric, I am using the EBS backed instance and not the instance store disk. – Sameer Manas May 9 '12 at 18:56

You can specify a larger initial root volume when you create an instance via the command line ec2 tools using the following example:

ec2-run-instances --key mykey --block-device-mapping /dev/sda1=:100 ami-3ac64a0a --region us-west-2 --group mygroup --instance-type m1.small -z us-west-2b

Then, after ssh'ing in, resize the drive:

sudo resize2fs /dev/xvda1

That should give you a m1.small instance in us-west-2b with a 100gb root volume.

share|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.