17:00:51 <hartsocks> #startmeeting VMwareAPI 17:00:52 <openstack> Meeting started Wed Sep 4 17:00:51 2013 UTC and is due to finish in 60 minutes. The chair is hartsocks. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:00:53 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 17:00:55 <openstack> The meeting name has been set to 'vmwareapi' 17:01:03 <hartsocks> greetings stackers! 17:01:06 <hartsocks> Who's online? 17:01:50 <hartsocks> \o 17:01:54 <tjones> yo 17:02:39 <hartsocks> Today's the last day to work before feature freeze. I'm sure folks are busy. 17:02:52 <vuil> hi 17:03:07 <danwent> vuil: perfect, was just planning on pinging you to ask if you were joining :) 17:03:10 <danwent> vui 17:03:48 <hartsocks> #link https://wiki.openstack.org/wiki/Meetings/VMwareAPI 17:04:00 <hartsocks> This is normally the part of the meeting where I ask about bugs. 17:04:09 <hartsocks> But I think everyone is pretty focused on reviews. 17:04:14 <hartsocks> Why don't we start there? 17:04:21 <danwent> seems smart 17:04:28 <hartsocks> #topic reviews 17:04:52 <hartsocks> I've been working off of the priority order list we built a while ago... 17:04:59 <hartsocks> #link https://review.openstack.org/#/c/30282/ 17:05:06 <hartsocks> This is the multiple-cluster review... 17:05:19 <hartsocks> It was rev-ed less than an hour ago. 17:05:45 <hartsocks> I'm waiting on the test results from CI before I re-review. 17:06:37 <danwent> yeah, garyk added some additional tests 17:06:44 <danwent> based on russell's request 17:06:54 <danwent> the good news is that we have the attention of two core reviewers 17:07:01 <hartsocks> *go team* 17:07:24 <smurugesan> Hey all, Sabari here 17:07:25 <tjones> yeah! and russell said it looks good other than tests 17:07:49 <hartsocks> smurugesan: hey glad you joined. 17:08:12 <hartsocks> Hopefully the last round of CI testing comes back clean and we can just have this merge. 17:08:17 <smurugesan> I couldn't make it for the last few weeks. Good to be here :D 17:08:18 <DarkSinclair> Greetings Dan, Shawn. When appropriate I'd like to make a request if possible. 17:08:48 <hartsocks> DarkSinclair: we'll have plenty of open-discussion time at the end this week. 17:08:55 <hartsocks> Next on my list 17:08:59 <hartsocks> #link https://review.openstack.org/#/c/40105/ 17:09:02 <hartsocks> Merged! 17:09:15 <hartsocks> #link https://review.openstack.org/#/c/40245/ 17:09:35 <dims> Hi all 17:09:36 <hartsocks> waiting to merge it looks like… has 2 +2's and an approve... 17:09:43 <hartsocks> dims: hey. 17:10:13 <hartsocks> #link https://review.openstack.org/#/c/41387/ 17:10:28 <hartsocks> Looks like it's waiting on an approve so it can merge. 17:10:49 <hartsocks> #link https://review.openstack.org/#/c/41600/ Merged! 17:11:02 <hartsocks> #link https://review.openstack.org/#/c/34903/ 17:11:09 <hartsocks> Deploy vCenter templates 17:11:09 <hartsocks> 17:11:24 <danwent> vuil: does it make sense for you to re-rev the template patch yourself? 17:11:27 <hartsocks> vui: looks like you have a −1 here? 17:12:37 <hartsocks> vuil: can you address this issue yourself? I think Kiran is travelling. 17:12:59 <vuil> sorry offscreen 17:13:14 <vuil> I am going to rerev the patch. 17:13:32 <hartsocks> vuil: thanks. 17:13:49 <hartsocks> vuil: if you need help figuring out the process, I'll be around late tonight. 17:13:59 <hartsocks> #link https://review.openstack.org/#/c/37659/ 17:14:12 <vuil> need to tease out what else needed to be done to the vm cloned from the template 17:14:28 <vuil> sure thanks Shawn. 17:14:39 <hartsocks> "Enhance VMware instance disk usage" is recently rev-ed. 17:14:51 <hartsocks> It's on my list to re-review next. 17:15:13 <hartsocks> #link https://review.openstack.org/#/c/37819/ 17:15:34 <hartsocks> sitting idle (that's the image clone stuff) not as important as the other features. 17:15:51 <hartsocks> The three bugs I was tracking are... 17:16:07 <hartsocks> merged except for: https://review.openstack.org/#/c/33100/ 17:16:17 <hartsocks> This is Sabari's .... 17:16:26 <hartsocks> "Fixes host stats for VMWareVCDriver" 17:16:29 <danwent> hartsocks: we should also talk about the second half of the cinder stuff 17:16:31 <smurugesan> yes, I think it's important to show the correct stats to the users 17:16:31 <hartsocks> This is a bug so it can go later. 17:16:50 <danwent> hartsocks: unless i missed that discussion already 17:16:58 <hartsocks> I mentioned them… 17:17:03 * hartsocks pulls up links 17:17:11 <danwent> including: https://review.openstack.org/#/c/43465/ 17:17:12 <hartsocks> nova-side cinder support 17:17:26 <danwent> not the nova side, there is an additional patch that is in cinder that has not really been on our radar 17:17:46 <danwent> it is owned by subbu 17:18:11 <danwent> but he is based in india, I believe, so if we get feedback from the core cinder team, we could use someone on our team to re-rev that 17:18:17 <hartsocks> ah. 17:18:40 <hartsocks> Adding it to my list. 17:19:16 <hartsocks> #link https://blueprints.launchpad.net/cinder/+spec/vmware-vmdk-cinder-driver 17:19:32 <hartsocks> Looks like ... 17:19:57 <hartsocks> first review merged, this second one didn't merge yet... 17:20:03 <danwent> hartsocks: exactly 17:20:19 <danwent> I just ping jgriffith and he said it was one of the items the team discussed at their meeting right before us 17:21:01 <hartsocks> danwent: normally, I sit in on jgriffith's meeting … I'll spend some time with this review myself then. 17:21:15 <hartsocks> Any other reviews we need to pay attention to? 17:21:38 <danwent> hartsocks: that is the only I noticed that wasn't on our list already 17:22:08 <hartsocks> #action vmwareapi team members review https://review.openstack.org/#/c/43465/ to support cinder team 17:23:24 <hartsocks> Hopefully, we don't have to apply for exception on any blueprint patches. I will send out a note after the 5th to the Mailing List if there's a significant miss. 17:23:41 <hartsocks> anything else on reviews folks want to bring up? 17:24:34 <hartsocks> Only a few more hours before the 5th. Thanks for staying on top of things! 17:24:41 <hartsocks> #topic bugs 17:24:44 <smurugesan> I would like some +1 s on the host stats patch that you mentioned last 17:24:53 <hartsocks> #undo 17:24:54 <openstack> Removing item from minutes: <ircmeeting.items.Topic object at 0x321a750> 17:25:14 <hartsocks> #action reviews for https://review.openstack.org/#/c/33100/ 17:25:25 <smurugesan> Also https://review.openstack.org/#/c/30628/ is a good change that needs to be in 17:26:10 <hartsocks> okay... 17:26:41 <hartsocks> Fortunately, bugs aren't under as much pressure as blueprints (bug fixes can be back ported). 17:27:07 <smurugesan> Good to know that Shawn! 17:27:26 <tjones> and we all love to backport ;-D 17:27:32 <hartsocks> #action give timely follow up on https://review.openstack.org/#/c/33100/ 17:27:39 <hartsocks> #undo 17:27:40 <openstack> Removing item from minutes: <ircmeeting.items.Action object at 0x332f510> 17:27:52 <hartsocks> #action give timely follow up on https://review.openstack.org/#/c/30628/ 17:28:02 <hartsocks> Okay... 17:28:17 <hartsocks> anything else we need to be aware of in front of the feature freeze? 17:28:47 <hartsocks> #link https://wiki.openstack.org/wiki/FeatureFreeze 17:29:17 <hartsocks> There is an exception procedure for feature freeze. I hope to not have to use it. 17:29:45 <hartsocks> #topic bugs 17:30:12 <hartsocks> Any bugs we need to be aware of? (that we aren't already?) 17:30:39 <hartsocks> going once... 17:30:55 <hartsocks> going twice... 17:30:57 <hartsocks> Okay... 17:31:26 <hartsocks> I think we're all properly focused on getting those critical blueprints through. 17:31:36 <hartsocks> #topic open discussion 17:31:47 <hartsocks> The floor is open… what's going on? 17:32:13 <hartsocks> DarkSinclair: ping 17:32:44 <DarkSinclair> I've had an internal request after an Audit to explore the opportunity of encrypting the vmware virtual center user's password in the nova-compute.conf file. Is this within teh realm of possiblity? 17:32:57 <hartsocks> heh. 17:33:02 <hartsocks> Possible? yes. 17:33:13 <hartsocks> At the moment you'll have to either… 17:33:26 <hartsocks> use a symmetric cipher 17:33:50 <hartsocks> … that is encrypt the password in the file, then decrypt it in memory right before Python SUDS uses it. 17:33:56 <hartsocks> or... 17:34:05 <hartsocks> we'll have to add SSO support to the driver. 17:34:13 <hartsocks> That's in a blue print for IceHouse... 17:34:22 * hartsocks digs around for blueprint link. 17:34:37 <hartsocks> #link https://blueprints.launchpad.net/nova/+spec/vmware-sso-support 17:34:45 <DarkSinclair> i was hoping the later ;) The next Audit request is around accountability and ownership, SSO support would alleviate both concerns. 17:34:52 <hartsocks> The vCenter SSO service starting at vSphere 5.1 ... 17:34:59 <hartsocks> allows for holder of key tokens. 17:35:11 <hartsocks> Yep. 17:35:46 <danwent> DarkSinclair: yeah, this is definitely something we want to work on 17:35:46 <hartsocks> I decided to not try and get SSO into Havana because we were running tight as it was. 17:35:50 <DarkSinclair> Is it possible to proxy authentication of the logged in Openstack user for vCenter authentication? 17:36:17 <hartsocks> hm... 17:36:18 <DarkSinclair> it's understood there's no chance for Havana. 17:36:27 <hartsocks> There's a few moving parts there... 17:36:29 <danwent> DarkSinclair: we'll probably provide a way to backport the change 17:36:41 <DarkSinclair> danwent: even better. 17:37:00 <danwent> but its unlikely that it will be backported to the official branch, unless we can spin it as a security issue 17:37:49 <hartsocks> We should probably have a conversation with some vSphere SSO experts on what the best thing to do is. 17:37:53 <DarkSinclair> My timeline is 6 months to a year, so it aligns with I-release 17:38:19 <hartsocks> There might be something clever we can do with a Keystone plugin either in keystone or in vCenter. 17:38:41 <hartsocks> When you say "proxy" you mean... 17:38:48 <DarkSinclair> does Keystone in havana better support Microsoft AD integration? (i havent checked yet. if so, that would be excellent.) 17:39:30 <hartsocks> to be honest I've not watched keystone closely this cycle. I'd have to dig in code. 17:39:55 <danwent> DarkSinclair: not sure, but exploring connections of our SSO and keystone is on our list to explore 17:39:56 <hartsocks> Would AD integration in keystone plus AD integration in vCenter (already there in 5.1) be enough to satisfy your req? 17:40:32 <hartsocks> That would mean there really wouldn't be much integration between keystone and vCenter SSO. 17:40:49 * russellb waves 17:40:51 <hartsocks> But… you would have to re-auth on both sides. 17:41:03 <hartsocks> russellb: hei hei! 17:41:31 <russellb> hartsocks: need anything while i'm around 17:41:32 <russellb> ? 17:41:58 <hartsocks> russellb: we're all anxiously watching those reviews… listed in the channel already. 17:42:02 <danwent> russellb: we really appreciate the work you've been doing on our reviews the past few days 17:42:42 <DarkSinclair> hartsocks: ideally all authentication is against individual users in AD. SSO is already setup and pulling from AD, if we could re-use that from Openstack and keystone, that'd be perfect 17:42:58 <russellb> danwent: np 17:43:05 <russellb> hartsocks: cool, reviewing what i can 17:43:16 <hartsocks> russellb: thanks. 17:43:32 <danwent> looks like jenkins is struggling a bit :) https://review.openstack.org/#/c/30282/ 17:44:11 <hartsocks> russellb: and I have been working on your IPC problem in the test harness a bit… :-) 17:44:23 <russellb> danwent: ah, that neutron thing again :-p 17:44:45 <russellb> though ideally the recheck wouldn't have said "no bug" 17:44:55 <russellb> need to reference the bugs so it's clear how much problems are affecting the gate 17:45:08 <danwent> russellb: agreed 17:45:39 <hartsocks> DarkSinclair: I'll record some of this conversation in the blueprint for discussion in the next design summit. 17:46:42 <danwent> russellb: we should be available on irc if anything pops up during the day 17:46:50 <hartsocks> russellb: if something/someone is stuck you can ping me 17:46:55 <DarkSinclair> hartsocks: thanks. 17:47:49 <russellb> hartsocks: danwent ok will do, will ping if i need updates on anything 17:48:01 <russellb> looking over the multiple cluster one again now 17:48:16 <hartsocks> russellb: thanks. 17:50:05 <hartsocks> Anything else folks need to discuss while we have people's attention? 17:51:27 <DarkSinclair> Will everything work w/ vSphere 5.5 ? 17:51:46 <DarkSinclair> aka: any known got-chas currently? 17:52:00 <hartsocks> DarkSinclair: we have had testing with 5.5 and so far we've not had anything come back to us. 17:52:38 <DarkSinclair> Perfect. 17:54:06 <hartsocks> DarkSinclair: we do have an open bug on using more than one dataCenter … but that's not a 5.5 specific thing… and we have a set of proposed fixes for the problem. They aren't blueprints (feature requests) so they've been lower priority for the last week or two. 17:54:42 <hartsocks> DarkSinclair: just a BTW thing. If you use one datacenter things work well. 17:55:52 <hartsocks> With that, I'll give people back a few minutes. 17:55:58 <hartsocks> Thanks for all your hard work. 17:56:02 <tjones> adios 17:56:15 <vuil> bye 17:56:19 <hartsocks> VMworld and this feature freeze at the same time was a lot to do! 17:56:33 <hartsocks> See you all next week. 17:56:44 <hartsocks> And we're in #openstack-vmware if you need us. 17:56:48 <hartsocks> #endmeeting