On 20.12.2018 23:32, John Ferlan wrote: > > > On 12/13/18 3:03 AM, Nikolay Shirokovskiy wrote: >> Signed-off-by: Nikolay Shirokovskiy <nshirokovskiy@xxxxxxxxxxxxx> >> --- >> docs/formatsnapshot.html.in | 7 +++++++ >> 1 file changed, 7 insertions(+) >> > > This probably should be merged with patch5 - nice to separate for review > though. Although, like patch4 it's not that important to me. > >> diff --git a/docs/formatsnapshot.html.in b/docs/formatsnapshot.html.in >> index fbbecfd..1357f53 100644 >> --- a/docs/formatsnapshot.html.in >> +++ b/docs/formatsnapshot.html.in >> @@ -235,6 +235,13 @@ >> at the time of the snapshot (<span class="since">since >> 0.9.5</span>). Readonly. >> </dd> >> + <dt><code>persistent/domain</code></dt> >> + <dd>Inactive domain configuration for active persistent domain. > > s/for/from an/ > >> + Such a domain have 2 distinct configs and here inactive is > > Tough read grammar wise. > >> + stored. It is different from <code>domain</code> which more >> + presisely keeps "inactive portion" of active config. > > precisely > >> + (<span class="since">since 5.0.0</span>). >> + </dd> > > > Hmmm... IIRC... isn't newDef the "next config"? Scour around for > Domain*Persistent and see what I mean. I think you have this description > backwards - hey I could be wrong, too... > > An active domain for which configuration specific changes have been made > will store both the "active" and "next config" in the domain object and > the <persistent> element will signify that config; whereas, the <domain> > element signifies the active domain configuration. Naming is known to be hard. For active domain you can think of this configs as active/persistent or active/next one(new one). I think @newDef name is choosen because @persistentDef would be misleading - inactive domains store its persistent domain in @def and not @newDef (however I would prefer the opposite - inactive domains storing its definition in @persistentDef and having @def = NULL). If done from scratch I would suggest to use: - <domain> for inactive/transient domains - <persistent><domain> and <transient><domain> for active persistent domains or even only last option for all cases. But now for backward compat we need some name. Well I thought explaining next/new def is more complicated then stating that persistent is only defined for active persistent domains and that is keeps the persistent config. Nikolay > > There is nothing different in the previous patch between the "<domain > ... </domain>" and the "<persistent> ... </persistent>"; however, I > would believe that if something changed as "config only" then it > would/should show up in that persistDom. > > John > >> <dt><code>cookie</code></dt> >> <dd>Save image cookie containing additional data libvirt may need to >> properly restore a domain from an active snapshot when such data >> -- libvir-list mailing list libvir-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/libvir-list