[
https://issues.apache.org/jira/browse/CLOUDSTACK-5547?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Nitin Mehta updated CLOUDSTACK-5547:
------------------------------------
Description:
Capacity reservation should be customizable at the vm level say with a flag in
the service offering. So when the flag is turned on and the vm is stopped we
reserve the capacity for the vm for stipulated period of time (configurable at
zone/cloud level). Currently the flag for turning on and off the reservation is
at cloud-wide and controlled through skip.counting.hours.
Also in CS, reserved capacity is tracked per host instead of per vm level and
this creates issues say when say a stopped vm is upgraded ans started where in
it eats into some other vms reserved capacity. We should move to a model of
tracking it per vm level.
was:
Capacity reservation should be customizable at the vm level say with a flag in
the service offering. So when the flag is turned on and the vm is stopped we
reserve the capacity for the vm for stipulated period of time (configurable at
zone/cloud level). Currently the flag for turning on and off the reservation is
at cloud-wide and controlled through skip.counting.hours.
Also in CS, reserved capacity is tracked per host instead of per vm level and
this creates issues say when say a stopped vm is upgraded etc. We should move
to a model of tracking it per vm level.
> Capacity reservation should be customizable at the vm level
> -----------------------------------------------------------
>
> Key: CLOUDSTACK-5547
> URL: https://issues.apache.org/jira/browse/CLOUDSTACK-5547
> Project: CloudStack
> Issue Type: Improvement
> Security Level: Public(Anyone can view this level - this is the
> default.)
> Affects Versions: 4.3.0
> Reporter: Nitin Mehta
>
> Capacity reservation should be customizable at the vm level say with a flag
> in the service offering. So when the flag is turned on and the vm is stopped
> we reserve the capacity for the vm for stipulated period of time
> (configurable at zone/cloud level). Currently the flag for turning on and off
> the reservation is at cloud-wide and controlled through skip.counting.hours.
> Also in CS, reserved capacity is tracked per host instead of per vm level and
> this creates issues say when say a stopped vm is upgraded ans started where
> in it eats into some other vms reserved capacity. We should move to a model
> of tracking it per vm level.
--
This message was sent by Atlassian JIRA
(v6.1.4#6159)
