On 6/17/13 7:43 PM, "Mice Xia" <mice_...@tcloudcomputing.com> wrote:
>Kelven, > >After the refactoring, will CS still restart HA enabled VM when it is >power off externally (e.g. using xencenter) or internally (user initiated >shutdown or crash)? If HA functionality is provided by external manager (i.e., vCenter), CS won't try to restart it, but if HA is provided by CloudStack, we will restore the legacy logic. The hook up with old HA manager (between HighAvailabilityManager/VirtualMachineManager) is in the wrap-up process > >Is seems the method HighAvailabilityManagerImpl .scheduleRestart() is not >called when VM's actual state is stopped while expected state is running. > > >Regards >Mice > >-----Original Message----- >From: Kelven Yang [mailto:kelven.y...@citrix.com] >Sent: Tuesday, June 18, 2013 5:21 AM >To: dev@cloudstack.apache.org >Subject: Re: [MERGE] Merge VMSync improvement branch into master > >Haven't created a patch yet, will do it soon after some last wrap-ups. > >Kelven > >On 6/17/13 12:03 PM, "John Burwell" <jburw...@basho.com> wrote: > >>Kelven, >> >>Did this patch get pushed to Review Board? If so, what is the URL? >> >>Thanks. >>-John >> >>On Jun 17, 2013, at 1:40 PM, Kelven Yang <kelven.y...@citrix.com> wrote: >> >>> Low level classes were tested in unit tests(MessageBus, Job >>>framework, Job dispatchers etc), interface layer changes are guarded >>>through matching the old semantics, but changes are tested manually, >>>we are planning to get this part of testing through BVT system after >>>we have re-based the latest master. >>> >>> Kelven >>> >>> On 6/17/13 10:01 AM, "Chip Childers" <chip.child...@sungard.com> wrote: >>> >>>> On Mon, Jun 17, 2013 at 04:59:00PM +0000, Kelven Yang wrote: >>>>> I'd like to kick off the official merge process. We will start the >>>>>merge process after the branch has passed necessary tests >>>>> >>>>> Kelven >>>> >>>> Can you share what testing is being run against the branch? >>> >> >
