Search Engine: Elastic

Article ID: 117548, created on Sep 27, 2013, last review on Jul 14, 2018

  • Applies to:
  • Operations Automation 5.4


A VPS Hardware Node capacities are shown as overused in the POA Provider Control Panel at Service Director > Virtuozzo Manager > VPS Hardware Nodes > select server > General > Capacities, the 'Available' field contains the negative value like on the screenshot below:

VPS HW capacity

Also, Provider is getting notifications from POA like "Usage of capacity 'Virtual Memory' on host '184' has been reached the limit (15399443), current consumption is 16700040.' for the hardware node in question.

In case of backups - it can manifest itself by backups being removed from node.


Every time POA notices a capacity overusage when checking it, it sends such notification. Not in all cases the notification about overused resources means the real problem with resources.

According to POA Provider's Guide (and the Parallels Virtuozzo Containers design) Provider may oversell the CPU and virtual memory resources, see more details at the following section in the docs - Limiting Number of VPSs per Hardware Node.

Therefore, in this particlar case as long as VPSs on the hardware node in question work stable such notifications from POA are harmless, they just notify you that the limit is reached.


POA calculates the Virtual Memory capacity and consumption this way:

Virtual Memory capacity = (hardware node RAM + swap) / 4KB pages

You may find RAM and swap values on the Statistics screen in properties of a particular VPS hardware node or on a hardware node itself.

Virtual Memory Consuption = sum of slmmemorylimit of VPS / 4KB for all VPSs on a hardware node

If some VPSs use the UBC resource allocation mode, they do not add to the consumption.

Diskspace for backups can be added through Top > Service Director > Virtuozzo Manager > VPS Hardware Nodes > <backup node> > Stored VPS backups > Configure.

If you want to get rid of notification consider adding more RAM, Diskspace or swap to the affected hardware node, so that consumption would be less than capacity, or migrate some VPSs to another servers. It will stop the notifications.

ac82ce33439a9c1feec4ff4f2f638899 caea8340e2d186a540518d08602aa065 5356b422f65bdad1c3e9edca5d74a1ae 2554725ed606193dd9bbce21365bed4e e12cea1d47a3125d335d68e6d4e15e07

Email subscription for changes to this article
Save as PDF