Re: [OSSTEST Nested PATCH v8 6/7] Compose the main recipe of nested test job

[Ian Campbell <ian.campbell@citrix.com>]

On Wed, 2015-04-22 at 12:04 +0100, Ian Campbell wrote:
> On Wed, 2015-04-22 at 08:38 +0000, Pang, LongtaoX wrote:
> > 
> > 
> > > -----Original Message-----
> > > From: Ian Campbell [mailto:ian.campbell@citrix.com]
> > > Sent: Tuesday, April 21, 2015 6:49 PM
> > > To: Pang, LongtaoX
> > > Cc: xen-devel@lists.xen.org; Ian.Jackson@eu.citrix.com; wei.liu2@citrix.com; Hu,
> > > Robert
> > > Subject: Re: [OSSTEST Nested PATCH v8 6/7] Compose the main recipe of nested
> > > test job
> > > 
> > > On Mon, 2015-04-13 at 17:19 -0400, longtao.pang wrote:
> > > > Signed-off-by: longtao.pang <longtaox.pang@intel.com>
> > > > ---
> > > > Changes in v8:
> > > > Change the patch order from 6 to 5.
> > > > ---
> > > >  sg-run-job |   11 +++++++++++
> > > >  1 file changed, 11 insertions(+)
> > > >
> > > > diff --git a/sg-run-job b/sg-run-job
> > > > index eae159d..2ca5ebf 100755
> > > > --- a/sg-run-job
> > > > +++ b/sg-run-job
> > > > @@ -299,6 +299,17 @@ proc run-job/test-pair {} {
> > > >  #    run-ts . remus-failover ts-remus-check         src_host dst_host +
> > > debian
> > > >  }
> > > >
> > > > +proc need-hosts/test-nested {} {return host}
> > > > +proc run-job/test-nested {} {
> > > > +    run-ts . = ts-debian-hvm-install + host + nestedl1
> > > > +    run-ts . = ts-nested-setup + host + nestedl1
> > > > +    run-ts . = ts-xen-install nested_l1
> > > > +    run-ts . = ts-host-reboot nested_l1
> > > > +    run-ts . = ts-debian-hvm-install nested_l1 nestedl2
> > > > +    run-ts . = ts-guest-stop nested_l1 nestedl2
> > > > +    run-ts . = ts-guest-destroy host nestedl1
> > > 
> > > Based on my comments to previous patches I think you'll want to use
> > > nestedl1 and nestedl2 consistently and never nested_l1/nested_l2.
> > > 
> > nestedl1 nestedl2 are hostname for L1 guest VM and L2 guest VM,
> > nested_l1 and nested_l2 are ident for L1 and L2.
> 
> Hostnames shouldn't be appearing in this script, I don't think,
> certainly not bare like that.
> 
> For the physical i.e. l0 case I think what happens is that
> need-hosts/foo ends up returning a bunch of $ident=$an_actual_host. i.e.
> srchost=$hosta and dsthost=$hostb.
> 
> I'm not sure how that would work in the nested world, but maybe you
> should be writing $ident=$guestname (i.e. nested_l1=nestedl1)?
> 
> Ian?

Ian is still plugging in his computers after moving desks. But having
spoken to him IRL it seems I'm not quite right above.

The arguments passed to ts-* in sg-run-job should _always_ be just a
bare indent. The ident=foo thing is a special case for standalone mode.

However, Ian did also say (and I agreed) that there really isn't any
reason why the ident and the hostname should differ here (or indeed the
guest name from l0's point of view, modulo the suffix osstest sometimes
adds).

Ian.