cloudstack-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Emil <emi...@gmail.com>
Subject RE: Problem with VolumeMigration, stuck after first migration‏
Date Wed, 14 Oct 2015 04:43:43 GMT
Ok, I will make some deep research.
I will let you know, a.s.a.p.
On Oct 14, 2015 00:21, "Somesh Naidu" <Somesh.Naidu@citrix.com> wrote:

> Ilya, the operation SEQ is based on the host (also referred to as agent).
> So, this concurrency isn't at the object level but at the host/agent level.
>
> In this case, I believe it is SSVM. Have Emil verify this from host table
> querying for the ID in the SEQ.
>
> Regards,
> Somesh
>
>
> -----Original Message-----
> From: ilya [mailto:ilya.mailing.lists@gmail.com]
> Sent: Tuesday, October 13, 2015 4:59 PM
> To: users@cloudstack.apache.org
> Subject: Re: Problem with VolumeMigration, stuck after first migration‏
>
> Hi Emil,
>
> I believe there are some restrictions on how migration is handled with
> vmware.
>
> Please explain from resource standpoint what you are running and how
> test is made..
>
> i.e. how many hosts and datastores in your cluster and how tests are made..
>
> Example:
> Test 1 Source: vm1, host 1, datastore1
> Test 1 Dest: vm1, host 1, datastore2
>
> Test 2 Source: vm1, host 2, datastore3
> Test 2 Dest: vm1, host 2, datastore4
>
>
>
> When you try to migrate, see of you can migrate vms that are on
> different host and datastore - but in the same cluster...
>
> I vaguely recall there was a limit of concurrent operations on single
> object...
>
> Regards
> ilya
>
>
> On 10/11/15 1:17 PM, Emil wrote:
> > Hello dear users,
> > we implemented cloudstack (4.5.2) in a private network, as hypervisor we
> > used VmWare (ESXi 5.5).
> >
> > The enviornment running perfectly except for one strange and annoying
> > problem.
> > When we are trying the migrate a volume it migrates very fine and we can
> > see it the vSphere Client,
> >
> > but !! when we are trying to make another migration it just enter the
> queue
> > (can see it with listAsyncJobs)
> > and it waits for the first migration to end, when the first migration
> ends
> > the second migration starting
> > Not only the second migration not running parallel or any third, fourth
> ...
> > ,
> > Any task we are trying to run enters the queue (deploy, snapshot,
> anything)
> > !
> >
> > We are tried to run every thing from the api, same results,
> > searched the logs for warning, but every thing seems to be OK,
> > it even wrote "job-165 entered for monitoring" (Sorry if it ain't
> accurate)
> > for the second migration or
> > any other task we are trying to execute
> >
> > I must admit that when we are trying to run anything else then migration,
> > it runs parallel (Taking Snapshots, Deployments)
> >
> > When I looked at the list of AsyncJobs (through the api), the jobstatus
> of
> > the new tasks (When first migration deployed) it valued as 0.
> >
> > Why it is stuck ? and not running parallel ? Can someone help here ?
> >
>

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