There should be a setting in the global config - Skip Config hours. This reserves the capacity for the stopped vm for this duration so that the user doesn't have to bother about the stopped vm not able to start again because of lack of capacity. You can set it to a smaller value or 0.
On 17/09/13 7:29 AM, "Mir Islam" <mis...@mirislam.com> wrote: >I saw that, but after reading about it on the wiki I was not sure if that >is what I was looking for. Memory multiplier is not even available. >Basically I want the resource (cpu and memory) to be available if >instance is not running. Currently, it does not matter what state (other >than destroyed) the instance is, a new instance can not be started. So no >way to over subscribe the resource from what I can gather. Is that >correct? > > >On Sep 16, 2013, at 6:28 PM, John Skinner wrote: > >> In global settings there is a setting for cpu over allocation >>multiplier, that will allow you to have more allocatable compute >>resources... I am not sure on memory however. >> >> >> ----- Original Message ----- >> From: "Mir Islam" <mis...@mirislam.com> >> To: users@cloudstack.apache.org >> Sent: Monday, September 16, 2013 3:41:45 PM >> Subject: CPU/Memory Allocation >> >> A simple question. Currently when I create new instance, available CPU >>and Memory resources go down. Even when the instance is not running. >>That prevents me to create new instance. Is there a setting somewhere >>where I can allocate/create new instances as long as the resources are >>not actually used? >> >> Thanks >> Mir >
