cloudstack-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ahmad Emneina <aemne...@gmail.com>
Subject Re: Asynchronous Request Handling
Date Wed, 20 Feb 2013 02:43:39 GMT
On Tue, Feb 19, 2013 at 6:19 PM, Chiradeep Vittal <
Chiradeep.Vittal@citrix.com> wrote:

>
>
> On 2/19/13 12:45 AM, "Asmita Vagyani" <Asmita.Vagyani@sigma-systems.com>
> wrote:
>
> >Hi all,
> >
> >We are using the cloudstack 4 APIs - deployVM and destroyVM.
> >Both the calls are asynchronous.
> >
> >We are trying to find a way how we can keep on polling the CS4 management
> >server for the Job status.
> >The issue is , the VM creation can take much time than expected.... is
> >there any time limit say x hours after which I can say I will will get
> >the Job Status as success/failure.
>
> Currently it is hard-coded as 1 day (24 hours). After this time the query
> will return 'non existent job' or some such.
>
>
could the globals parameters vm.op.* or job.expire.minutes help guarantee a
fail time in these scenarios?



>  >Because, VM creation process cannot go on forever.
> >
> >If, during the process of creation if I want to abort this VM creation
> >process, can I do it?
>
> Looking at the VM state machine, if the state is in 'Starting', the
> end-user has no way of forcing a state transition to say 'Destroyed' by
> sending the 'destroyVM' command.
>
> But I could be wrong about this. I have not tried this.
>
>
I believe this is the current behavior. You cant operate on vm's when
theyre in transition states.


>  >
> >Thanks and Regards.
> >
> >Asmita Patil Vagyani.
> >
>
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message