13:03:23 <claudiub> #startmeeting hyper-v
13:03:24 <openstack> Meeting started Wed Feb 15 13:03:23 2017 UTC and is due to finish in 60 minutes.  The chair is claudiub. Information about MeetBot at http://wiki.debian.org/MeetBot.
13:03:25 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
13:03:29 <openstack> The meeting name has been set to 'hyper_v'
13:03:48 <claudiub> hello
13:11:21 <claudiub> so, it seems there isn't anyone today.
13:11:42 <claudiub> just going to note a few things quickly.
13:12:37 <claudiub> #topic miscellaneous topics
13:13:49 <claudiub> just a small fyi, we have made a demo with Kubernetes with Windows and Linux containers + OVS & OVN, if anyone is curious
13:14:07 <claudiub> #link Kubernetes with Windows and Linux containers using OVS & OVN: https://www.youtube.com/watch?v=N02K7ZGnVXI
13:15:13 <claudiub> another quick note is that we won't have a Hyper-V meeting next week, as most of us will be at the PTG in Atlanta. I'll have to send a mail to the ML about this as well.
13:16:37 <claudiub> and if anyone is going there and wishes to meet up to discuss Hyper-V / Windows related topics and future plans, you can drop me a line on IRC or an email, and we'll schedule a meeting.
13:18:46 <claudiub> regarding the nova Hyper-V driver, we are currently working on a major refactor regarding cold resize / migration.
13:19:44 <claudiub> it is needed in order to fix a bug: https://bugs.launchpad.net/os-win/+bug/1663238
13:19:44 <openstack> Launchpad bug 1663238 in OpenStack Compute (nova) "Hyper-V driver destroys and recreates the VM on cold migration / resize" [Medium,Triaged]
13:20:52 <claudiub> basically, the nova Hyper-V driver destroys the VM on source and recreates it on destination, which can cause the PCI device IDs to change, causing undesired effects in some cases
13:21:58 <claudiub> and also, it is needed for shielded VMs as well, in order to allow them to cold migrate or cold resize. If the VMs are destroyed, their TPM data is lost, which is bad.
13:24:08 <claudiub> the VM can be imported on the destination node instead of being recreated, but we have to consider updating all the VM resources in the case of a resize (PCI devices, QoS policies, vNUMA topologies, etc.). Which means that a few patches to os-win are needed first.
13:26:38 <claudiub> currently considering the RemoteFX resize scenario. after this, it should be pretty much done.
13:27:08 <claudiub> that's about it for now.
13:27:17 <claudiub> #endmeeting