Ask Your Question
0

If I boot from a Volume (cinder), why appears as I'm using Compute's Local Disk Usage?

asked 2016-02-22 18:52:35 -0500

sacha-m gravatar image

Scenario:

  • One Controller Node (Controller)
  • One Storage Node using Cinder with LVM with 900GB for storage (Storage)
  • One Compute node with 100GB as local disk basically for OS usage (Compute)

In Horizon I create a volume with type, one of the types I created and "lives" in Storage Node (let's call it lvm-1), and an cloud image as Volume Source. 800GB as Volume Size. Then, I launch an instance from this volume, with a flavor that uses 1 CPU, 2GB RAM and 700GB of disk. If I go to Admin -- System -- Hypervisors -- Hypervisor Summary appears all that 700GB in Compute's local Storage usage, so it appears as that hypervisor has allocate more disk than it has, but AFAIK that disks "lives" in Storage node. Am I right? If I am, why Horizon is showing the opposite?

edit retag flag offensive close merge delete

2 answers

Sort by » oldest newest most voted
1

answered 2016-02-23 01:30:25 -0500

Hi, copying exact words from this post: http://cloudarchitectmusings.com/2013/11/18/laying-cinder-block-volumes-in-openstack-part-1-the-basics/

  1. A volume is created through the cinder create command. This command creates an LV into the volume group (VG) “cinder-volumes.”
  2. The volume is attached to an instance through the nova volume-attach command. This command creates a unique iSCSI IQN that is exposed to the compute node.
  3. The compute node, which runs the instance, now has an active iSCSI session and new local storage (usually a /dev/sdX disk).
  4. Libvirt uses that local storage as storage for the instance. The instance get a new disk, usually a /dev/vdX disk.

Here another post about cinder flows: http://j-griffith.blogspot.com.es/2015/07/volume-attach-code-flow-in-cinder.html

Regards

edit flag offensive delete link more

Comments

Ok, and in that case Horizon shouldn't add this /dev/sdX/ that get from the storage node to the total amount of local disk that has compute nodes? Because appears that has only 100GB (in my example) instead of 800GB.

sacha-m gravatar imagesacha-m ( 2016-02-23 06:21:31 -0500 )edit

I just check, and the compute node that runs the instance doesn't have this new /dev/sdX disk

sacha-m gravatar imagesacha-m ( 2016-02-23 07:19:57 -0500 )edit

That name is the name inside the instance, not the name in compute node. If you login to your instances, there should be a device called /dev/vd* or something similar.

Eduardo Gonzalez gravatar imageEduardo Gonzalez ( 2016-02-23 10:09:53 -0500 )edit

Its only 100 GB because is provisioned in thin mode, that means than when you request a 800 GB, a real size of the volume is not 800, is a minor value, commonly near to the data is inside the volume.

Eduardo Gonzalez gravatar imageEduardo Gonzalez ( 2016-02-23 10:11:20 -0500 )edit

Yes, appears as /dev/vda, as expected. My point is that disk lives for real in Cinder, which is not in Compute node but in Storage node, so.. why appears as those 800GB were "taken" from compute node. That's it.

sacha-m gravatar imagesacha-m ( 2016-02-24 07:21:14 -0500 )edit
0

answered 2016-05-20 07:38:53 -0500

ZaphodB gravatar image

This is a bug in nova-compute which is currently being addressed. See my answer here: https://ask.openstack.org/en/question/65854/incorrect-advertisement-of-block-storage-total-and-free-space/?answer=92450#post-id-92450 (https://ask.openstack.org/en/question...)

edit flag offensive delete link more

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer

Get to know Ask OpenStack

Resources for moderators

Question Tools

2 followers

Stats

Asked: 2016-02-22 18:52:35 -0500

Seen: 1,751 times

Last updated: Feb 23 '16