Re: Livepatching and Xen Security

[Lars Kurth <lars.kurth@citrix.com>]

On 18/05/2017 17:53, "Ian Jackson" <ian.jackson@eu.citrix.com> wrote:

>George Dunlap writes ("Livepatching and Xen Security"):
>> # Executive summary
>
>I am completely in agreement with your analysis and your conclusions.

Me too. I am not sure though whether we need a vote or lazy consensus.

For Credit2 (see 
https://lists.xenproject.org/archives/html/xen-devel/2016-11/msg00171.html)
 we used a code patch removing Experimental from KCONFIG, which implicitly
led to active confirmation by project leadership members via Acked-by
tags. In other words, we executed the equivalent of a vote.

From a process perspective
https://xenproject.org/governance.html#lazyconsensus should be sufficient
because we are applying a process, not changing one. In addition lazy
consensus decisions can only be overturned if the project leadership
agrees collectively to do so, because the decision is too important for
lazy consensus. As every leadership member is on the list, there would be
ample opportunity to raise objections.

My gut feeling though is that Leadership Team members and Security Team
members should probably actively agree/disagree to proposals related to
whether a feature is supported and thus security supported to avoid any
future issues. George already expressed an implicit +1 (by making the
proposal) and Ian an explicit +1.

Regards
Lars

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel