From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ian Campbell Subject: Re: [OSSTEST Nested PATCH v7 4/6] Add new script to custmize nested test configuration Date: Wed, 1 Apr 2015 09:58:33 +0100 Message-ID: <1427878713.2115.233.camel@citrix.com> References: <1427497587-22884-1-git-send-email-longtaox.pang@intel.com> <1427497587-22884-5-git-send-email-longtaox.pang@intel.com> <1427811214.2115.147.camel@citrix.com> <86C3224E41A7434B904EC364302132D80E487DED@SHSMSX101.ccr.corp.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <86C3224E41A7434B904EC364302132D80E487DED@SHSMSX101.ccr.corp.intel.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: "Pang, LongtaoX" , Ian.Jackson@eu.citrix.com Cc: "Hu, Robert" , "wei.liu2@citrix.com" , "xen-devel@lists.xen.org" List-Id: xen-devel@lists.xenproject.org On Wed, 2015-04-01 at 08:45 +0000, Pang, LongtaoX wrote: > > As it happens I was rebasing that series this morning but due to other issues > > I've not managed to run it yet. Once I've managed to at least smoke test I'll CC > > you on the repost. > > > OK. What's more, for the below codes which is used for starting > 'osstest-confirm-booted' script to confirm whether L1 is fully booted > after reboot it. I think it's necessary here. > +target_cmd_root($l1, < + wget -O overlay.tar $url > + tar -xf overlay.tar -C / > + rm overlay.tar -f > + update-rc.d osstest-confirm-booted start 99 2 . > +END In my distro series I also have some patches refactoring the overlay stuff, which would mean you could reuse that. http://article.gmane.org/gmane.comp.emulators.xen.devel/224433 I'll CC you on that one too. I don't think there would be any harm in adding those overlays for all guests and enabling the initscript, but Ian may disagree or know something which I don't. > > > +my $l2_disk_mb = 20000; > > > +my $l2_lv_name = 'nestedl2'; > > > +my $vgname = $l0->{Name}; > > > +$vgname .= ".$c{TestHostDomain}" if ($l0->{Suite} =~ m/lenny/); > > > +target_cmd_root($l0, < > > + lvremove -f /dev/${vgname}/${l2_lv_name} ||: > > > + lvcreate -L ${l2_disk_mb}M -n $l2_lv_name $vgname > > > + dd if=/dev/zero of=/dev/${vgname}/${l2_lv_name} count=10 > > > + xl block-attach $l1->{Name} > > > +/dev/${vgname}/${l2_lv_name},raw,sdb,rw > > > > This attach won't be persistent over domain destroy, is that expected? > > > We need create a logical volume in L0 and attach it to L1 for L2 VM > installing. Before creating, it will remove the logical volume with > the same "$l2_lv_name" that has already existed. > Yes, after destroyed L1 VM, the logical volume will still exist inside > L0, but there is no harm, I think. OK. Ian./