16:00:43 #startmeeting hyper-v 16:00:43 Meeting started Tue Aug 5 16:00:43 2014 UTC and is due to finish in 60 minutes. The chair is primeministerp. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:00:45 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:00:47 The meeting name has been set to 'hyper_v' 16:00:53 Hi everyone 16:01:06 virmitio: hi tim 16:01:17 ociuhandu: hi tavi 16:01:28 hi all 16:01:36 primeministerp: hello 16:01:45 luis_fdez: hi luis 16:02:15 hi there 16:02:19 alexpilotti: hey there 16:02:29 quite the quorum today 16:02:34 no vijay 16:02:52 so some updates on active devel work 16:03:01 #topic cinder smb3 support 16:03:06 alexpilotti: you're up 16:03:38 alexpilotti: smb3 status? 16:03:42 sure 16:03:57 so as usual we’re at full steam here 16:04:05 alexpilotti: as expected 16:04:23 on teh CInder side we already committed the SMB patches since a while 16:04:35 fetching them 1sec 16:04:42 alexpilotti: manilla as well? 16:04:52 in case some additional volunteers would fancy a review ;-) 16:05:34 #link https://review.openstack.org/#/c/106046/ 16:05:49 #link https://review.openstack.org/#/c/106057/ 16:06:04 alexpilotti: awesome thx 16:06:25 so we’ve been asked to rebase teh SMB work on top of another patch 16:06:36 nasically an NFS refactoring that reduces some common code 16:06:53 and we have to wait for that patch to merge before ours are taken into considerations 16:07:07 I’ve been assured that this should happen soon 16:07:10 ok 16:07:22 on that note 16:07:29 Cinder anyway is not Nova, so usually there are no review bandwidth issues that affect us 16:07:52 #topic cinder ci iscsi/smb3 16:08:10 ociuhandu: where's the automation stand for the cinder tests 16:08:32 primeministerp: we have the image prepared 16:08:36 primeministerp: and the scripts 16:08:47 primeministerp: we’re waiting to clear out the licensing issues 16:08:55 ociuhandu: which 16:09:01 ociuhandu: I'll deal w/ that 16:09:02 now 16:09:04 windows storage server license 16:09:24 in the lab we currently should have any issue 16:09:27 i have keys 16:09:33 if to be used 16:09:39 er to be used 16:09:46 on a bigger scale 16:09:49 if we need a kms 16:09:52 or something else 16:10:02 I can look into that today 16:10:06 and get it resolved today 16:10:11 primeministerp: alsoi take into account that the license key will be activated for each test 16:10:21 ociuhandu: fair enough 16:10:42 primeministerp: that can be an issue we were waiting for an answer on (and also the actual license to be used) 16:10:45 ociuhandu: can you send me an email with the explict requirements 16:11:16 ociuhandu: needed for the license and any other reqs 16:11:19 as far as i know the key might have issues if it is being activated more than a certain number of times 16:11:23 ociuhandu: i'll push some buttens today 16:11:30 primeministerp: sure thing 16:11:34 ociuhandu: thx 16:11:43 ociuhandu: also can we use the same image for both? 16:12:11 primeministerp: both what? 16:12:19 ociuhandu: smb and iscsi 16:12:23 primeministerp: iscsi and smb? 16:12:28 yes 16:12:32 primeministerp: yes we can 16:12:42 perfecto 16:12:58 ociuhandu: also can we deploy on iron using ironic yet? 16:13:01 ;) 16:13:19 primeministerp: we also have a few nodes set aside for these tests 16:13:26 ociuhandu: which? 16:13:36 not iron but cinder CI 16:13:50 ociuhandu: can we use iron? 16:13:51 so we will not mix the kvm nodes for cinder with the ones for nova and neutron, for the moment 16:14:10 i.e. how much ram can we squeeze in 16:14:21 primeministerp: at the moment we are using kvm, we can discuss about iron later on 16:14:46 well, from our tests we can squeeze the windows image in about 2 to 3 Gb 16:14:50 ociuhandu: given our ram issues iron would be better 16:15:15 ociuhandu: and we can hard set those to pxe 16:15:36 ociuhandu: if we wanted to use ironic 16:15:42 primeministerp: given the existing hardware I would not reinstall them so often 16:16:01 ociuhandu: well neither would i but it would give us a cleaning cycle 16:16:03 ;) 16:16:04 and also I would not rely on one windows per node as if that node fails we have bigger problems 16:16:12 the nodes we have are with 32 Gb ram 16:16:22 so quite a lot of instances per node 16:16:44 ociuhandu: we can probably deal w/ numbers 16:16:55 ociuhandu: i can't give you ram until RR is done 16:16:59 ociuhandu: anyway 16:17:06 alexpilotti: nova?