RE: [Xen-devel] RE: "xm create" take more time to return.

["Zhang, Yang Z" <yang.z.zhang@xxxxxxxxx>]

Hi keir
Our testing is base on xm now . And we know xl will be official supported by 
xen 4.1, so should we move to xl now or after xen 4.1 release?

BTW: we had investigated xl several months ago, and there were some problems 
with it(I have send a mail to discuss it before), but I am not sure whether 
those bugs are fixed or not.

best regards
yang


> -----Original Message-----
> From: Keir Fraser [mailto:keir.xen@xxxxxxxxx] On Behalf Of Keir Fraser
> Sent: Wednesday, November 24, 2010 5:03 PM
> To: Zhang, Yang Z; xen-devel@xxxxxxxxxxxxxxxxxxx
> Cc: Ian Jackson
> Subject: Re: [Xen-devel] RE: "xm create" take more time to return.
> 
> Noone cares that much about xm any more. Reproduce with xl create and you
> might get some takers from the maintainers of libxl.
> 
> 
> On 24/11/2010 03:14, "Zhang, Yang Z" <yang.z.zhang@xxxxxxxxx> wrote:
> 
> > No one care about the stress case? Or this issue only happen to me?
> >
> > best regards
> > yang
> >
> >
> >> -----Original Message-----
> >> From: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx
> >> [mailto:xen-devel-bounces@xxxxxxxxxxxxxxxxxxx] On Behalf Of Zhang, Yang
> Z
> >> Sent: Friday, November 19, 2010 9:47 AM
> >> To: xen-devel@xxxxxxxxxxxxxxxxxxx
> >> Cc: Keir Fraser
> >> Subject: [Xen-devel] "xm create" take more time to return.
> >>
> >> Hi all
> >> Now I am running stress test against xen-unstable. We have one case that
> >> repeated to create/destroy guest one thousand times. I can see it is always
> >> successful to create/destroy guest at begin of the testing(about first 
> >> three
> >> hundreds). But after create/destroy guest about three hundreds, we start
> to
> >> see the testing will fail sometime. And at ending of the testing, it always
> >> fail to
> >> create guest.
> >> After investigation, we found after many times create/destroy guest. The
> >> return from "xm create guest.config" was very slowly. It may take more
> than 1
> >> minutes to return. But guest still successful to create when it returned.
> >> With
> >> "top", I found the xenstored take about 70% cpu utilization when run "xm
> >> create". So I think there would be something wrong with xenstored after
> >> create/destroy guest many many times.
> >> Anyone have suggestion?
> >>
> >> best regards
> >> yang
> >>
> >>
> >>
> >> _______________________________________________
> >> Xen-devel mailing list
> >> Xen-devel@xxxxxxxxxxxxxxxxxxx
> >> http://lists.xensource.com/xen-devel
> >
> > _______________________________________________
> > Xen-devel mailing list
> > Xen-devel@xxxxxxxxxxxxxxxxxxx
> > http://lists.xensource.com/xen-devel
> 


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel