yahoo-eng-team team mailing list archive

[Aymen Frikha <aymen.frikha@xxxxxxxxxxxxx>]

Public bug reported:

Hello,

We are facing in a production openstack Mitaka a situation where the
used RAM (reported by Nova) is greater than the available RAM (reported
by Nova) and the ram allocation ratio is set  to 0.9.

Used ram: 624.5 GB Available:503.8 GB

The real available memory in the compute is 70GB.

Is it something we need to worry about ( Even if all instances in the compute start using their real allocated memory) ?
We are not able to reproduce this behavior in a test environment.

Environment:

ii  nova-common                          2:13.1.4-0ubuntu4.2                        all          OpenStack Compute - common files
ii  nova-compute                         2:13.1.4-0ubuntu4.2                        all          OpenStack Compute - compute node base
ii  nova-compute-kvm                     2:13.1.4-0ubuntu4.2                        all          OpenStack Compute - compute node (KVM)
ii  nova-compute-libvirt                 2:13.1.4-0ubuntu4.2                        all          OpenStack Compute - compute node libvirt support
ii  python-nova                          2:13.1.4-0ubuntu4.2                        all          OpenStack Compute Python libraries


Hypervisor: Libvirt + KVM

Storage: Ceph

Netowkring: Neutron + linuxbridge

** Affects: nova
     Importance: Undecided
         Status: New

-- 
You received this bug notification because you are a member of Yahoo!
Engineering Team, which is subscribed to OpenStack Compute (nova).
https://bugs.launchpad.net/bugs/1805887

Title:
  Used RAM greater than Available RAM with ram allocation ratio is set
  to 0.9

Status in OpenStack Compute (nova):
  New

Bug description:
  Hello,

  We are facing in a production openstack Mitaka a situation where the
  used RAM (reported by Nova) is greater than the available RAM
  (reported by Nova) and the ram allocation ratio is set  to 0.9.

  Used ram: 624.5 GB Available:503.8 GB

  The real available memory in the compute is 70GB.

  Is it something we need to worry about ( Even if all instances in the compute start using their real allocated memory) ?
  We are not able to reproduce this behavior in a test environment.

  Environment:

  ii  nova-common                          2:13.1.4-0ubuntu4.2                        all          OpenStack Compute - common files
  ii  nova-compute                         2:13.1.4-0ubuntu4.2                        all          OpenStack Compute - compute node base
  ii  nova-compute-kvm                     2:13.1.4-0ubuntu4.2                        all          OpenStack Compute - compute node (KVM)
  ii  nova-compute-libvirt                 2:13.1.4-0ubuntu4.2                        all          OpenStack Compute - compute node libvirt support
  ii  python-nova                          2:13.1.4-0ubuntu4.2                        all          OpenStack Compute Python libraries

  
  Hypervisor: Libvirt + KVM

  Storage: Ceph

  Netowkring: Neutron + linuxbridge

To manage notifications about this bug go to:
https://bugs.launchpad.net/nova/+bug/1805887/+subscriptions