14:00:20 #startmeeting sahara 14:00:21 Meeting started Thu Aug 9 14:00:20 2018 UTC and is due to finish in 60 minutes. The chair is tellesnobrega. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:22 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:24 The meeting name has been set to 'sahara' 14:00:54 o/ 14:01:04 o/ 14:01:38 lets start 14:01:42 #topic News/Updates 14:01:52 still working on S3 testing 14:03:18 not much new to report from me 14:03:41 from me neither, mostly reviews and will prepare cycle highlights and cut RC1 today 14:04:18 I will be on PTO next week, so no sahara-tests release for now, but it's independent anyway 14:04:49 ok 14:05:14 I'd like t ocomplete this S3 support, but it requires some changes in sahara-scenario to support S3 parameters 14:06:26 that's why I asked jeremyfreudberg about how how to pass the parameters for S3 in https://review.openstack.org/#/c/589645/ 14:07:04 cool 14:07:23 do we have any specific concerns for RC1? 14:07:38 for rc1, i'd like to include my s3 doc patch 14:08:04 the only other real patch we have is the fixing anti-affinity 14:08:12 it may have to cause an rc2 14:08:22 yes, no time for it now 14:08:32 I will review the doc after this meeting 14:08:36 cool 14:09:34 zchkun_, how have you been? 14:10:35 I'm sorry, no progress on my side 14:10:54 it was delayed by other things. 14:11:03 zchkun_, no problem 14:15:10 topics for today? 14:15:45 sorry 14:15:57 I don't have specific stuff to discuss, I think tosky might have something 14:17:06 I, uh, don't 14:17:23 I have few jobs-related task after the branching 14:18:28 i have some comments on boot from volume which we can discuss 14:19:21 jeremyfreudberg, go aheaddd 14:19:29 #topic boot-from-volume 14:20:13 yes, so in rocky we added the ability to specify that an instance should boot from volume, as a node group template option 14:20:43 we have another set of ngt options for additional attached volumes, which get used to back hdfs specifically 14:21:02 for these attached volumes, we give a lot more configuration options: 14:21:38 volumes_per_node, volumes_size, volumes_availability_zone, volume_type, volume_local_to_instance, volume_mount_prefix 14:22:09 some of those aren't relevant to boot from volume 14:22:48 volumes_per_node, volumes_size (for bootable volume, it comes from the flavor), volume_mount_prefix (a bootable volume has its own way of mounting, i think) 14:23:01 but the other ones are potentially useful 14:23:39 so we should have a brief discussion on how the interface will be for that 14:23:40 true, we can definitely take a look and see how we can improve it 14:24:17 basically the question is, should we create new options like boot_from_volume_type, boot_form_volume_az, boot_from_volume_locality, etc 14:24:47 or should the current set of options for attached volumes also affect the bootable volume 14:25:10 my own thought, is that we should create new options 14:25:32 the user might have different requirements for the boot volume (for the os) versus for the attached volume (for hdfs specifically) 14:25:46 I'm not entirely sure, I will have to dig a little more to make a decision 14:25:58 yes, it will be in stein anyway, so we can think 14:26:05 it makes sense to have separated 14:26:44 tosky-- any preliminary thoughts? 14:27:15 but I want to double check, so we don't have too much stuff and complicate things 14:27:36 tellesnobrega: yes, i agree with that too 14:28:24 but it is good that you brought it up, we can take some time to look it up, understand the pros and cons 14:28:42 and if we don't decide prior to PTG we have time there to talk about it 14:29:09 probably better separately 14:30:09 cool 14:30:40 that's it from me 14:33:02 I'm done as well 14:34:00 nothing else from me too 14:34:07 tellesnobrega, reminder to make the ptg etherpad 14:34:28 jeremyfreudberg, it is on my queue 14:34:38 probably get to it tomorrow 14:35:14 thanks everyone 14:35:24 we get 25 minutes back 14:35:34 #endmeeting