Well, you should be able to do metering. The network and disk stats are basically counters like you'd get from an SNMP call to a switch. You have to calculate the diff between reads.
On Tue, Mar 17, 2015 at 4:45 PM, Nux! <n...@li.nux.ro> wrote: > Oh right. > Was hoping I can do some sort of metering with them, as there are no usage > network stats in a basic/adv+sg zone. > > Thanks for the clarification. I'll have to resort to something else. > > > Lucian > > -- > Sent from the Delta quadrant using Borg technology! > > Nux! > www.nux.ro > > ----- Original Message ----- >> From: "Marcus" <shadow...@gmail.com> >> To: dev@cloudstack.apache.org >> Sent: Tuesday, 17 March, 2015 22:26:49 >> Subject: Re: Reset instance "Statistics" > >> They are stored in memory. There's no history. For KVM there was/is a >> bug in cpu stats where it grabs an average since uptime. >> >> https://issues.apache.org/jira/browse/CLOUDSTACK-8247?filter=-1 >> >> It should be do-able to reset this concurrent hashmap to zero via api >> call if someone wants to add that feature. >> >> On Tue, Mar 17, 2015 at 1:48 PM, Somesh Naidu <somesh.na...@citrix.com> >> wrote: >>> Lucian, >>> >>> These values get collected every minute (controlled by global config >>> vm.stats.interval) from the hypervisor. >>> >>> These values do not seem to be persisted in DB. I guess there won't be a >>> need to >>> reset these. >>> >>> Somesh >>> CloudPlatform Escalations >>> Citrix Systems, Inc. >>> >>> >>> -----Original Message----- >>> From: Nux! [mailto:n...@li.nux.ro] >>> Sent: Tuesday, March 17, 2015 2:31 PM >>> To: dev@cloudstack.apache.org >>> Subject: Reset instance "Statistics" >>> >>> Hello, >>> >>> Anyone knows how often these values get reset or if at all? >>> http://img.nux.ro/fK7-Selection_016.png >>> >>> I'm interested in tracking some of these values, but on a monthly basis or >>> so. >>> >>> Lucian >>> >>> -- >>> Sent from the Delta quadrant using Borg technology! >>> >>> Nux! >> > www.nux.ro
