| *** matsuhashi has joined #openstack-trove | 00:23 | |
| *** mayfield_ has joined #openstack-trove | 01:08 | |
| *** hub_cap_ has joined #openstack-trove | 01:08 | |
| *** esmute_ has joined #openstack-trove | 01:12 | |
| *** dmakogon_ has quit IRC | 01:13 | |
| *** mayfield has quit IRC | 01:13 | |
| *** esmute has quit IRC | 01:13 | |
| *** hub_cap has quit IRC | 01:13 | |
| *** esmute_ is now known as esmute | 01:13 | |
| *** dmakogon_ has joined #openstack-trove | 01:20 | |
| *** matsuhashi has quit IRC | 01:24 | |
| *** matsuhashi has joined #openstack-trove | 01:26 | |
| *** jasonb365 has joined #openstack-trove | 01:31 | |
| *** nosnos has joined #openstack-trove | 01:31 | |
| *** jasonb365 has quit IRC | 01:39 | |
| *** jasonb365 has joined #openstack-trove | 01:40 | |
| *** yidclare has quit IRC | 01:45 | |
| *** hub_cap_ is now known as hub_cap | 01:49 | |
| hub_cap | man looks like i dc'd... a netsplit maybe? /me shrugs | 01:55 |
|---|---|---|
| *** eghobo has quit IRC | 02:19 | |
| *** khyati has joined #openstack-trove | 02:59 | |
| *** khyati has quit IRC | 03:02 | |
| *** eghobo has joined #openstack-trove | 03:07 | |
| *** haomaiwang has joined #openstack-trove | 03:17 | |
| *** matsuhashi has quit IRC | 03:19 | |
| *** jasonb365 has quit IRC | 03:19 | |
| *** nosnos has quit IRC | 03:34 | |
| *** sbfox has joined #openstack-trove | 03:54 | |
| *** haomaiw__ has joined #openstack-trove | 03:56 | |
| *** matsuhashi has joined #openstack-trove | 03:57 | |
| *** haomaiwang has quit IRC | 03:58 | |
| *** nosnos has joined #openstack-trove | 04:02 | |
| *** michael-yu has joined #openstack-trove | 04:03 | |
| *** michael-yu has quit IRC | 04:04 | |
| *** michael-yu has joined #openstack-trove | 04:06 | |
| *** matsuhashi has quit IRC | 04:39 | |
| *** matsuhashi has joined #openstack-trove | 04:43 | |
| *** jasonb365 has joined #openstack-trove | 04:53 | |
| *** SushilKM has joined #openstack-trove | 05:37 | |
| *** jasonb365 has quit IRC | 05:37 | |
| *** nosnos_ has joined #openstack-trove | 05:53 | |
| *** nosnos has quit IRC | 05:53 | |
| *** matsuhashi has quit IRC | 05:54 | |
| *** matsuhashi has joined #openstack-trove | 05:55 | |
| openstackgerrit | Jenkins proposed a change to openstack/trove: Imported Translations from Transifex https://review.openstack.org/82721 | 06:24 |
| *** flaper87|afk is now known as flaper87 | 07:02 | |
| *** matsuhashi has quit IRC | 07:08 | |
| *** matsuhashi has joined #openstack-trove | 07:08 | |
| *** nosnos_ has quit IRC | 07:09 | |
| *** nosnos has joined #openstack-trove | 07:10 | |
| *** matsuhashi has quit IRC | 07:22 | |
| *** yogesh has joined #openstack-trove | 07:26 | |
| *** matsuhashi has joined #openstack-trove | 07:28 | |
| *** michael-yu has quit IRC | 07:30 | |
| *** sbfox has quit IRC | 07:32 | |
| *** michael-yu has joined #openstack-trove | 07:36 | |
| *** michael-yu has quit IRC | 07:37 | |
| *** matsuhashi has quit IRC | 07:46 | |
| *** matsuhashi has joined #openstack-trove | 07:47 | |
| *** timfreund has quit IRC | 07:53 | |
| *** yogesh has quit IRC | 07:55 | |
| *** eghobo has quit IRC | 08:29 | |
| *** yogesh has joined #openstack-trove | 08:43 | |
| *** yogesh has quit IRC | 08:44 | |
| *** RockKuo_TW_PC has joined #openstack-trove | 08:48 | |
| *** RockKuo_TW has quit IRC | 08:52 | |
| *** haomaiw__ has quit IRC | 09:13 | |
| *** haomaiwa_ has joined #openstack-trove | 09:13 | |
| *** haomaiw__ has joined #openstack-trove | 09:32 | |
| *** haomaiwa_ has quit IRC | 09:35 | |
| *** timfreund has joined #openstack-trove | 09:42 | |
| *** matsuhashi has quit IRC | 11:26 | |
| *** nosnos has quit IRC | 11:27 | |
| *** demorris has joined #openstack-trove | 11:50 | |
| *** saju_m has joined #openstack-trove | 11:51 | |
| *** saju_m has quit IRC | 12:00 | |
| *** saju_m has joined #openstack-trove | 12:06 | |
| *** saju_m has quit IRC | 12:08 | |
| openstackgerrit | Denis M. proposed a change to openstack/trove: Fix create call for security group rules https://review.openstack.org/83768 | 12:17 |
| openstackgerrit | Denis M. proposed a change to openstack/python-troveclient: Fix create call for security group rules https://review.openstack.org/84115 | 12:18 |
| openstackgerrit | Denis M. proposed a change to openstack/trove: Add stack_id for future usage https://review.openstack.org/67873 | 12:20 |
| openstackgerrit | Denis M. proposed a change to openstack/trove: Support floating IP association at provisioning https://review.openstack.org/45075 | 12:20 |
| openstackgerrit | Denis M. proposed a change to openstack/trove: Allow db instance conditional logging https://review.openstack.org/63789 | 12:20 |
| openstackgerrit | Denis M. proposed a change to openstack/trove: Allow log files audit https://review.openstack.org/64302 | 12:21 |
| openstackgerrit | Denis M. proposed a change to openstack/python-troveclient: Allow log files audit https://review.openstack.org/64303 | 12:21 |
| openstackgerrit | Denis M. proposed a change to openstack/trove: Track security group provisioned by heat https://review.openstack.org/71040 | 12:21 |
| openstackgerrit | Denis M. proposed a change to openstack/trove: Add heat.template for Percona datastore https://review.openstack.org/76536 | 12:21 |
| openstackgerrit | Denis M. proposed a change to openstack/trove: Add heat.template for Redis datastore https://review.openstack.org/76538 | 12:21 |
| openstackgerrit | Denis M. proposed a change to openstack/trove: Add point in time recovery https://review.openstack.org/77222 | 12:21 |
| openstackgerrit | Denis M. proposed a change to openstack/python-troveclient: Add point in time recovery https://review.openstack.org/77223 | 12:21 |
| openstackgerrit | Denis M. proposed a change to openstack/trove: Support full backup/restore for Cassandra https://review.openstack.org/77551 | 12:21 |
| *** pdmars has joined #openstack-trove | 12:23 | |
| openstackgerrit | Denis M. proposed a change to openstack/trove: Refactoring strategies implementation https://review.openstack.org/77899 | 12:26 |
| openstackgerrit | Denis M. proposed a change to openstack/trove: Refactor mysql and redis apps https://review.openstack.org/80020 | 12:26 |
| *** pdmars has quit IRC | 12:26 | |
| *** pdmars has joined #openstack-trove | 12:30 | |
| *** demorris has quit IRC | 12:30 | |
| *** mrsnivvel has joined #openstack-trove | 12:36 | |
| *** Barker has joined #openstack-trove | 12:44 | |
| *** radez_g0n3 is now known as radez | 12:48 | |
| tenaglia | Do you know what is the status of RHEL-like support for Trove? | 12:51 |
| *** Barker has quit IRC | 12:54 | |
| *** dmakogon_ is now known as denis_makogon | 12:55 | |
| denis_makogon | tenaglia, please elaborate your question | 12:56 |
| tenaglia | denis_makogon: I understand it was not very clear. The last time we've checked the only platform on which Trove was able to run was Ubuntu. We'd like to use RHEL-like distros so I was wondering what is the current state. | 12:57 |
| tenaglia | python-troveclient had some Ubuntu-specific dependencies. | 12:58 |
| *** grapex has joined #openstack-trove | 13:03 | |
| denis_makogon | tenaglia, which deps. are ubuntu-specific ? | 13:04 |
| tenaglia | don't remember now. | 13:05 |
| tenaglia | maybe I should just give it a second shot and report later. | 13:05 |
| tenaglia | it was just a preliminary check. | 13:05 |
| denis_makogon | from my perspective, trove could work on fedora, i don't see so much difference between RHEL and fedora | 13:10 |
| *** grapex has quit IRC | 13:11 | |
| *** mrsnivvel has quit IRC | 13:20 | |
| *** blueyes has joined #openstack-trove | 13:44 | |
| *** blueyes has quit IRC | 13:46 | |
| *** blueyes has joined #openstack-trove | 13:46 | |
| *** zigo has quit IRC | 13:50 | |
| *** grapex has joined #openstack-trove | 13:57 | |
| *** robertmy_ has joined #openstack-trove | 13:58 | |
| *** zigo has joined #openstack-trove | 13:58 | |
| tenaglia | denis_makogon: indeed not much difference | 13:59 |
| tenaglia | thanks it's enough to re-start working at it. | 13:59 |
| denis_makogon | tenaglia, you're welcome | 14:02 |
| *** blueyes has quit IRC | 14:05 | |
| *** blueyes has joined #openstack-trove | 14:06 | |
| *** rwsu has joined #openstack-trove | 14:06 | |
| *** jasonb365 has joined #openstack-trove | 14:15 | |
| *** demorris has joined #openstack-trove | 14:15 | |
| *** blueyes has quit IRC | 14:18 | |
| *** michael-yu has joined #openstack-trove | 14:19 | |
| *** jcru has joined #openstack-trove | 14:23 | |
| *** amytron has joined #openstack-trove | 14:24 | |
| *** michael-yu has quit IRC | 14:26 | |
| *** tanisdl has joined #openstack-trove | 14:29 | |
| *** jmontemayor has joined #openstack-trove | 14:34 | |
| *** kevinconway has joined #openstack-trove | 14:37 | |
| *** kevinconway has quit IRC | 14:38 | |
| *** kevinconway has joined #openstack-trove | 14:38 | |
| *** freyes has joined #openstack-trove | 14:54 | |
| *** SushilKM has quit IRC | 14:56 | |
| *** Barker has joined #openstack-trove | 14:57 | |
| *** sbfox has joined #openstack-trove | 14:59 | |
| imsplitbit | I have sent out links to the instance metadata blueprint and spec | 15:01 |
| imsplitbit | core reviews blueprints on Mondays yes? | 15:01 |
| *** Barker has quit IRC | 15:01 | |
| *** jasonb365 has quit IRC | 15:02 | |
| *** Barker has joined #openstack-trove | 15:04 | |
| denis_makogon | imsplitbit, saw your email, we can talk about it today | 15:05 |
| *** demorris has quit IRC | 15:18 | |
| *** demorris has joined #openstack-trove | 15:18 | |
| hub_cap | denis_makogon: imsplitbit core will decide what bps to look at in the morning meeting | 15:20 |
| denis_makogon | hub_cap, i thought we're reviewing scheduled BPs, isn't it ? | 15:22 |
| hub_cap | it depends | 15:23 |
| hub_cap | if any random person puts a bp on then not necessarily | 15:23 |
| hub_cap | and the cor ewill choose what, based on what people put | 15:23 |
| hub_cap | there will still be a priority | 15:23 |
| denis_makogon | get it, i thought that core are going review what they want, independently from the long-term schedule | 15:25 |
| hub_cap | im not sure what that means | 15:27 |
| denis_makogon | let me elaborate abit, contributors have one week (from Mon. to Mon, legal time range) to schedule discussion (one or more BP) and all scheduled BPs would be discussed within 1h, and almost nothing can't change this order | 15:30 |
| hub_cap | no i said the opposite | 15:32 |
| hub_cap | put your bp up, we will choose which ones to look at | 15:32 |
| *** sbfox has quit IRC | 15:32 | |
| hub_cap | and you have as much time as you want to file a blueprint | 15:32 |
| hub_cap | there is no legal time range for that | 15:32 |
| hub_cap | im not sure what that means either | 15:32 |
| hub_cap | if it takes u 6 wks to finish your blueprint, then thas fine | 15:32 |
| denis_makogon | i mean legal time range - its the time when you should schedule your BP for discussion, if you schedule it like one or 2h ago, it'll be hard to review it | 15:36 |
| *** sbfox has joined #openstack-trove | 15:39 | |
| denis_makogon | when do we plan to start BP meeting ? | 15:44 |
| *** freyes has quit IRC | 15:45 | |
| *** khyati has joined #openstack-trove | 15:47 | |
| *** eghobo has joined #openstack-trove | 15:59 | |
| *** ViswaV has joined #openstack-trove | 16:02 | |
| *** jmontemayor_ has joined #openstack-trove | 16:03 | |
| *** jmontemayor has quit IRC | 16:06 | |
| denis_makogon | hub_cap, saw you're email | 16:23 |
| denis_makogon | hub_cap, is it true =( ? | 16:23 |
| hub_cap | heh, i did, in fact, send that email denis_makogon :) | 16:24 |
| hub_cap | i think 3 or 4 ptls are stepping down too | 16:24 |
| hub_cap | its a painful, relentless job | 16:24 |
| denis_makogon | i see | 16:25 |
| *** SushilKM has joined #openstack-trove | 16:28 | |
| *** jmontemayor_ has quit IRC | 16:29 | |
| *** jmontemayor has joined #openstack-trove | 16:31 | |
| denis_makogon | hub_cap, when do we plan to start meeting ? | 16:37 |
| hub_cap | im pretty sure it was at 11pt | 16:38 |
| denis_makogon | 11 pst ? | 16:42 |
| hub_cap | i think so denis_makogon | 16:45 |
| hub_cap | as in, 1hr15m from now | 16:45 |
| *** RockKuo_TW_PC has quit IRC | 16:46 | |
| denis_makogon | probably, i'll go home, will try not to be late | 16:46 |
| denis_makogon | hub_cap, thanks for converting the actual time =) | 16:46 |
| hub_cap | npnp | 16:47 |
| *** harlowja_away is now known as harlowja | 16:48 | |
| *** demorris has quit IRC | 16:52 | |
| amrith | hub_cap: how many minutes from now is "one hour from now" in PST ;) | 17:07 |
| *** michael-yu has joined #openstack-trove | 17:09 | |
| *** rwsu has quit IRC | 17:11 | |
| *** demorris has joined #openstack-trove | 17:13 | |
| hub_cap | lol | 17:14 |
| *** shivamshukla has joined #openstack-trove | 17:17 | |
| *** eghobo has quit IRC | 17:27 | |
| *** eghobo has joined #openstack-trove | 17:28 | |
| ViswaV | hub_cap: Hat tip for a job well done so far as the Trove PTL. Hope the baton passes on to equally capable hands. | 17:29 |
| *** khyati has quit IRC | 17:29 | |
| amrith | +1 | 17:29 |
| *** yidclare has joined #openstack-trove | 17:29 | |
| *** michael-yu has quit IRC | 17:29 | |
| hub_cap | thx guys | 17:30 |
| hub_cap | ill still be around to give whomever that is hell | 17:31 |
| hub_cap | and all of you | 17:31 |
| hub_cap | my -2 is still strong :) | 17:31 |
| ViswaV | :) | 17:31 |
| hub_cap | in reality i just want to be invited to the HP core contributor dinners in the OS conf's | 17:31 |
| *** mattgriffin has joined #openstack-trove | 17:41 | |
| * juice is tearing up | 17:44 | |
| juice | core reviewers - here is a blueprint that I would like to discuss today if there are open slots https://blueprints.launchpad.net/trove/+spec/trove-guest-private-network | 17:46 |
| *** amcrn has joined #openstack-trove | 17:46 | |
| *** rustlebee is now known as russellb | 17:47 | |
| *** SushilKM has quit IRC | 17:47 | |
| *** tanisdl has quit IRC | 17:47 | |
| vipul | hub_cap, SlickNik: are there BPs that we have on the docket to review | 17:49 |
| SlickNik | vipul: the agenda is here: https://wiki.openstack.org/wiki/Meetings/TroveMeeting#Agenda_for_Mar._31_.28blueprint_meeting.29 | 17:51 |
| *** mattgriffin has quit IRC | 17:52 | |
| *** mattgriffin has joined #openstack-trove | 17:54 | |
| *** yogesh has joined #openstack-trove | 17:56 | |
| *** denis_makogon_ has joined #openstack-trove | 17:58 | |
| amrith | for the BP review meeting are we using meeting-alt? or do we stay here? | 17:59 |
| *** denis_makogon has quit IRC | 17:59 | |
| *** denis_makogon_ is now known as denis_makogon | 17:59 | |
| *** dmakogon_ has joined #openstack-trove | 17:59 | |
| denis_makogon | amrith, we're here | 17:59 |
| hub_cap | denis_makogon: plz make your blueprints look more like juice's | 18:00 |
| SlickNik | I think we're planning on staying here. | 18:00 |
| amrith | ok, thx | 18:00 |
| hub_cap | we talked about the format already, and im not sure we can move forward w/ yours until you fix the way they look :) | 18:00 |
| denis_makogon | hub_cap, which one ? | 18:00 |
| hub_cap | both | 18:00 |
| amrith | I'll pour myself a stiff drink ;) | 18:00 |
| hub_cap | lets start w/ juices | 18:00 |
| hub_cap | its nice | 18:01 |
| juice | ready when you are | 18:01 |
| amrith | yes, I have the juice of a potato, the juice of a grape, juice of barley seed, ... | 18:01 |
| hub_cap | lol amrith | 18:01 |
| hub_cap | nice | 18:01 |
| juice | making me thirsty | 18:01 |
| hub_cap | im pretty sure thats all alcohol ;) | 18:01 |
| amrith | come on over and we can enjoy together. speaking of which, are you coming to attend today | 18:02 |
| amrith | alcohol, shiver the thought | 18:02 |
| amrith | I wouldn't touch the devils brew | 18:02 |
| hub_cap | naw i cant, im stuck in berkley till wed morn | 18:02 |
| juice | did the gun go off? | 18:02 |
| SlickNik | Yeah, go for it | 18:02 |
| amrith | ok, will miss you. others who are here please let me know ... percona live in Hyatt at Santa Clara is where its all happening | 18:02 |
| juice | I'll give you all a few minutes to skim the content | 18:02 |
| SlickNik | Agenda: https://wiki.openstack.org/wiki/Meetings/TroveMeeting#Agenda_for_Mar._31_.28blueprint_meeting.29 | 18:03 |
| amrith | who's driving? | 18:03 |
| juice | basically the premise here is to leverage neutron functionality in a way that we can create isolated trove instances | 18:03 |
| SlickNik | Right now we're talking about "Trove Managed Instances on Private Network" | 18:03 |
| juice | isolated in the sense that there are nearly inaccessible to intrusion with the exception of their db port | 18:03 |
| amrith | thx | 18:04 |
| juice | and they have no exposure to fiddling with from a nova stand point | 18:04 |
| juice | there is a behind the scenes tenant owned by the provider which owns all trove nova instances | 18:04 |
| *** khyati_ has joined #openstack-trove | 18:04 | |
| juice | from Trove's perspective, the accounting of which tenant owns which node is still being handled | 18:04 |
| juice | there is also a behind the scenes network which connects the Trove instances with the RabbitMQ server | 18:05 |
| hub_cap | yea so my beef here | 18:06 |
| juice | the user provides a network preferably their private network to the Trove Intance Create which allows them to access the Trove Instance from their App Server or if public from the internet | 18:06 |
| hub_cap | is that we are defining _how_ trove is installed | 18:06 |
| juice | you're from texas so I am expecting a lot of beef | 18:06 |
| hub_cap | im not sure that other projects are doing that | 18:06 |
| hub_cap | lol juice | 18:06 |
| juice | I think in this case we are providing the option to run trove in this fashion | 18:06 |
| hub_cap | do other projects lke nova make this the case? | 18:06 |
| hub_cap | well your bp says "by default" | 18:07 |
| juice | the idea is that it will support this use case but also allows for the default situation | 18:07 |
| amrith | juice, do I undrestand correctly that this means that we will no longer be able to address a trove instance as an underlying "nova instance" | 18:07 |
| hub_cap | ok the default is the dev env w/ 1 network | 18:07 |
| juice | hub_cap: yes the default will be the same as it is now | 18:07 |
| hub_cap | woah thsi is not cool juice , the complete trove mgmt | 18:07 |
| denis_makogon | hub_cap, even now we can assign more than one nics for the instance on provisioning | 18:08 |
| hub_cap | thats going too far | 18:08 |
| juice | but then through configuration and some plugin like code you can run the setup in this way | 18:08 |
| hub_cap | can u explain juice what "system-based tenatn will own all trove guest instances"? | 18:08 |
| hub_cap | is this just for the network? | 18:08 |
| juice | nope it's for the instances as well | 18:08 |
| hub_cap | i disagree completely on this | 18:09 |
| hub_cap | its not up to us to decide that | 18:09 |
| juice | however the billing/usage will still report the end user tenant id | 18:09 |
| hub_cap | this feels like the "hp" way is dictated into trove OS :) | 18:09 |
| ViswaV | +1 hub_cap | 18:09 |
| juice | hub_cap: even if it is just a configurable deployment | 18:10 |
| amrith | juice: how does one get access to things like query log, mysql general log, etc? if we can't access the underlying server? | 18:10 |
| vipul | I think this is simply proposing another remote.py impl that doesn't re-use the original token | 18:10 |
| hub_cap | vipul: we can make it an option | 18:10 |
| hub_cap | im fine w/ that | 18:10 |
| vipul | and giving you the option via cfg to choose that over the delegated auth method | 18:10 |
| juice | choice usually does not suggest dictate | 18:10 |
| hub_cap | but none o fthis can be default | 18:10 |
| hub_cap | id recommend updating the bp to define that none of this will be default. but options | 18:10 |
| juice | hub_cap: it will not be the default configuration | 18:10 |
| hub_cap | heh it doesnt appear as such based on the bp.. so jus tadd some verbiage and im cool w it | 18:11 |
| vipul | the benefit of this is also for those running Trove on a 'shared' nova -- it allows the instances to be hidden from the Nova CLI | 18:11 |
| juice | hub_cap: sorry if I suggested that somewhere. it certainly wasn't the intention | 18:11 |
| hub_cap | it makes sense to have this if u want | 18:11 |
| hub_cap | vipul: but we also talked w/ nova coutnless times about managed vms | 18:11 |
| SlickNik | Yeah, sounds to me like these can be deployment options, but shouldn't be the default trove config. | 18:11 |
| hub_cap | there is more than one way to solv ethis | 18:11 |
| hub_cap | w/o mucking w/ trove | 18:11 |
| hub_cap | just dont let people do things "managed" by trove | 18:11 |
| vipul | hub_cap: sure, but that's a long pole :) | 18:11 |
| hub_cap | or savanna or any other things | 18:11 |
| hub_cap | sure but itll help everyone | 18:11 |
| hub_cap | and it wont add extra code to trove | 18:11 |
| grapex | juice: I feel like I'm still wrapping my head around it, but I wonder if this could be two blueprints. One is the ability to configure Trove to allow another tenant to control server creation in Nova on a users behalf | 18:12 |
| ViswaV | also juice: the diagram https://wiki.openstack.org/wiki/File:Trove-private-network-overview.jpg does not show swift service as part of the trove management customer net. Don't we need that for backups/restore? | 18:12 |
| grapex | I wonder if that should be split out into it's own bp? | 18:12 |
| juice | grapex: I suppose it could be split into two | 18:13 |
| denis_makogon | one for the multi-tenancy, another for the networking, i guess | 18:13 |
| * grapex fist pumps | 18:13 | |
| juice | grapex: however, you can't do the networking one without the management user | 18:13 |
| juice | grapex: easy now easy | 18:14 |
| grapex | juice: Maybe it's dependent on it then? | 18:14 |
| hub_cap | exactly | 18:14 |
| amrith | in that case tehre has to be at least a third BP | 18:14 |
| juice | viswav: for backups, the swift account will still be tied to the end user | 18:14 |
| juice | amrith: what? :) tell me you're kidding | 18:15 |
| denis_makogon | juice, agreed, containers it's the actual user quota | 18:15 |
| amrith | for how you will allow multiple guests to share a VM and how you'll allow users access to files (like logs) | 18:15 |
| vipul | So one thing to keep in mind is.. there is an issue today where the instances will not have access to the control plane (unless you poke a hole) -- and when we go to Neutron, and the end user dictates the network to attach, we will lose control over whether that instance also has a management network | 18:15 |
| amrith | so if I understand this correctly, you want to lock down the instance | 18:15 |
| amrith | and not allow the user to access the underlyign Nova | 18:15 |
| amrith | and you want to also have >1 guest on a VM | 18:15 |
| amrith | if there's already a plan for how you get access to stuff like logs from the machine, then it should be part of this effort | 18:16 |
| amrith | no? | 18:16 |
| juice | amrith: correct, correct, wrong | 18:16 |
| amrith | ok | 18:16 |
| juice | the logging issue still exists here. whatever denis_makagon is working on should work here | 18:16 |
| amrith | fine, so long as it's been thought through | 18:16 |
| ViswaV | juice: I meant doesn't the swift service need to be accessible via the trove management net interface on the guest VM? Or is it expected to be accessible via the customer specified net during creation? | 18:16 |
| vipul | juice: if we don't go with a 'trove tenant' approach, how will we attach a management network to the instance | 18:16 |
| juice | viswav: I will look into the backup/swift issue | 18:16 |
| SlickNik | amrith: I don't think the bp proposes > 1 guest on a VM. | 18:17 |
| *** konetzed has quit IRC | 18:17 | |
| amrith | isn't that what shared Nova means (from what was said above) | 18:17 |
| grapex | vipul: You're saying that if Trove doesn't own the Nova server, attaching the network is impossible? | 18:17 |
| amrith | yes, I don't see that in the bp but that was being said above? | 18:17 |
| amrith | called multi-tenancy etc., | 18:17 |
| juice | vipul: we can still attach the instance to the management network behind the scenes. the issue is that the customer since they own the instance can go into nova and detach the port | 18:17 |
| vipul | amrith: Shared nova means you have a public nova endpoint.. that Trove deploys on top of | 18:17 |
| juice | :( | 18:17 |
| *** mattgriffin has quit IRC | 18:17 | |
| amrith | hmm | 18:18 |
| vipul | amrith: it means folks are doing a nova boot as well as trove create on the same nova | 18:18 |
| hub_cap | id like us to talk to nova before we go much further here tho | 18:18 |
| hub_cap | can u do some diligence here vipul/juice to discuss w/ nova core | 18:18 |
| hub_cap | every time ive brought htis up they like the idea | 18:18 |
| grapex | I also don't think, strictly speaking, you have to make Trove "own" the instances for this to be useful | 18:18 |
| hub_cap | of a "managed" vm | 18:18 |
| vipul | juice: Since we're using the Tenant's token, there will be lack of privileges for that tenant to know about mgmt network | 18:19 |
| grapex | for example, if no one has access to the Nova API you're talking to it it won't matter. | 18:19 |
| grapex | Just to name a highly theoretical example. :) | 18:19 |
| hub_cap | i think we _should_ keep everything in the tenant personally | 18:19 |
| SlickNik | vipul / amrith: Yeah, I think a more apt description would be "public nova" rather than "shared nova" | 18:19 |
| ViswaV | 'trove owned instances (single system tenant)" may also limit the freedom of an org to apply different QOS criteria based on tenant info (provision project1's trove instances on a cheaper set of Hyps or a different datacenter etc) | 18:19 |
| denis_makogon | hub_cap, agreed, multi-tenancy is the complicated part with spread services | 18:20 |
| juice | vipul: let me try to see if that is an issue. I think you can still provide a port from another network and attach it to the instance | 18:20 |
| *** shivamshukla has quit IRC | 18:20 | |
| amrith | LAGCHECK | 18:20 |
| *** tanisdl has joined #openstack-trove | 18:20 | |
| amrith | oops | 18:20 |
| vipul | grapex: Yes, it's not impossible. assuming you make the mgmt network public or give Neutron Admin to end users | 18:20 |
| amrith | need new password, sorry | 18:20 |
| juice | vipul: you may not be able to provide a net-id but I think you can provide a port that was created by the owner of that network | 18:21 |
| vipul | the issue is a end user wants a Private DB -- how do we also manage that private DB that lives in a private network | 18:21 |
| denis_makogon | juice, yes | 18:21 |
| hub_cap | amrith: yes yer lagging lol | 18:21 |
| amrith | * need barley juice | 18:22 |
| hub_cap | vipul: we can have a mgmt user to do _that_ | 18:22 |
| hub_cap | the validation on neuttron | 18:22 |
| vipul | hub_cap: Yea, so if we attach the mgmt network as a separate step in the boot process, via admin user that owuld work | 18:22 |
| hub_cap | but i dont think that alone justifies making everything in trove talk over a single user | 18:23 |
| juice | the end issue being that there is a lot of control in the users hands when they have access to nova | 18:23 |
| vipul | hub_cap: agreed i think it's two BPs.. | 18:23 |
| hub_cap | for sure, discrete issue ju | 18:23 |
| juice | putting the "managed" feature in Nova would be great | 18:23 |
| hub_cap | lol | 18:23 |
| hub_cap | my latency is terible........ | 18:23 |
| denis_makogon | mine too | 18:23 |
| juice | I think the latency is in my head | 18:23 |
| vipul | let's forget about the Trove tenant / ownership issue.. that can come later | 18:24 |
| hub_cap | juice: hehe you arent on a remote machine irc'ing | 18:24 |
| grapex | I think everyone could get behind having an option for Trove to manage the Nova servers. | 18:24 |
| juice | summary: i will split this up in two, talk to nova core to see if this is a feature they can provide. | 18:24 |
| vipul | grapex: yea that could come as an optional impl of remote.py later | 18:24 |
| juice | if this is not a feature they can provide, what next? | 18:24 |
| denis_makogon | juice, first in nova, then in heat, do not forget about it | 18:25 |
| juice | grapex: vipul: we do something to this effect today with the technique vipul mentioned above | 18:25 |
| juice | denis_makagon: heat?! seriously? | 18:25 |
| denis_makogon | juice, and only =) | 18:25 |
| grapex | juice: The heat is on. | 18:25 |
| juice | grapex: it's on the street | 18:26 |
| hub_cap | ++ i like this now | 18:26 |
| hub_cap | sry to ruin yoru world juice but arent u glad we talked ;) | 18:26 |
| denis_makogon | if nova/heat would be able to assign resources on given tenant it'll be fantastic | 18:27 |
| hub_cap | this is the whole point of these | 18:27 |
| juice | hub_cap: yeah this is all good | 18:27 |
| juice | better now than later | 18:27 |
| juice | early feedback is what this is all about so I think it's working | 18:28 |
| hub_cap | ++ | 18:28 |
| hub_cap | we have a late addition to th elist to fyi | 18:28 |
| vipul | Ok looks like we've beat this thing dead lol | 18:28 |
| vipul | next one? | 18:28 |
| denis_makogon | probably yes | 18:29 |
| hub_cap | vipul: lol | 18:29 |
| hub_cap | everyone refersh their page | 18:29 |
| juice | yeah I can't take anymore - Next! | 18:29 |
| SlickNik | I think so, we've got clear next steps for this bp. | 18:29 |
| juice | kinda fun though - I feel like I just ran through a gauntlet with you guys and your padded batons - "American Gladiator" style | 18:30 |
| SlickNik | *an instance metadata bp shows up* | 18:30 |
| hub_cap | denis_makogon: where are teh blueprints for your "other" things? | 18:30 |
| hub_cap | juice: dibs on the tenis ball gun | 18:31 |
| denis_makogon | hub_cap, https://blueprints.launchpad.net/trove/+spec/point-in-tyme-recovery | 18:31 |
| hub_cap | i just see your wiki links | 18:31 |
| hub_cap | plz fix | 18:31 |
| denis_makogon | hub_cap, https://blueprints.launchpad.net/trove/+spec/volume-snapshot | 18:31 |
| juice | hub_cap: oh man I forgot about that. good one | 18:31 |
| hub_cap | also imsplitbit plz fix yers | 18:31 |
| vipul | are we doing in order? | 18:31 |
| hub_cap | we can go w/ network attr mgmt now sure | 18:31 |
| hub_cap | this one is not in the proper format | 18:32 |
| SlickNik | Actually, the other 2 networking bps are both not in the proper format. | 18:32 |
| denis_makogon | hub_cap, done, page fixed | 18:32 |
| hub_cap | thx denis_makogon | 18:32 |
| hub_cap | yea SlickNik niether is | 18:32 |
| hub_cap | i thought they were both denis_makogon but one is annashen_ 's | 18:32 |
| SlickNik | They also talk about very similar things. | 18:33 |
| denis_makogon | mine for networks is only one | 18:33 |
| hub_cap | denis_makogon: plz remove the wiki page from there | 18:33 |
| hub_cap | the wiki page _should_ be linked from the bp | 18:33 |
| hub_cap | again | 18:33 |
| hub_cap | plz see juices | 18:33 |
| hub_cap | if its alreayd linked there is no reason to put it on the wiki page | 18:33 |
| vipul | which one are you guys looking at? https://blueprints.launchpad.net/trove/+spec/network-manager-spec ? | 18:34 |
| hub_cap | i wnat to kick them both to next wk | 18:34 |
| hub_cap | and let annashen_ and denis_makogon work out the differences | 18:34 |
| hub_cap | and fix their bp styleing | 18:34 |
| annashen_ | hub_cap, sure | 18:34 |
| SlickNik | dmakogon_ / annashen_ : Can you please update your bps to follow https://wiki.openstack.org/wiki/TroveBlueprint? | 18:34 |
| hub_cap | ++ | 18:34 |
| hub_cap | any blueprint not in that style will be immed kicked | 18:34 |
| * hub_cap is mean | 18:35 | |
| denis_makogon | btw, first BP also was with the bad style | 18:35 |
| annashen_ | SlickNick, hub_cap, i heard you both | 18:35 |
| hub_cap | annashen_: thank you | 18:35 |
| hub_cap | ok moving on? | 18:35 |
| SlickNik | annashen_: Thanks! | 18:35 |
| SlickNik | yup. | 18:35 |
| vipul | denis_makogon: first one we reviewed was perfect :) | 18:35 |
| hub_cap | yup | 18:36 |
| SlickNik | dmakogon_: take a look at https://wiki.openstack.org/wiki/Trove/trove-managed-instances (linked from the bp) | 18:36 |
| hub_cap | it was exactly what we are looking for | 18:36 |
| hub_cap | if they are not like thta, we will kick them | 18:36 |
| denis_makogon | SlickNik, bp isn't wiki page | 18:36 |
| hub_cap | weve already asked for them to be in thsi format | 18:36 |
| denis_makogon | let's move on | 18:36 |
| hub_cap | no denis_makogon | 18:36 |
| hub_cap | lets discuss this | 18:36 |
| vipul | BP should link to a wiki page | 18:36 |
| hub_cap | cuz youre wrong | 18:36 |
| hub_cap | the bp linksto the wiki | 18:36 |
| hub_cap | just like weve been doing forever | 18:36 |
| *** mattgriffin has joined #openstack-trove | 18:36 | |
| hub_cap | Set the url for this specification | 18:37 |
| hub_cap | that has always been there | 18:37 |
| hub_cap | and we have always use dit | 18:37 |
| hub_cap | lets do this one https://blueprints.launchpad.net/trove/+spec/trove-metadata | 18:37 |
| hub_cap | since its also in the correct format | 18:37 |
| hub_cap | any objections | 18:38 |
| hub_cap | ? | 18:38 |
| vipul | +1 let's do it | 18:38 |
| hub_cap | frck imsplitbit this is nice mang | 18:39 |
| hub_cap | *frack | 18:39 |
| cp16net | ahh damn PATCH? | 18:39 |
| cp16net | lolz | 18:39 |
| hub_cap | cp16net: yes | 18:39 |
| * SlickNik is reading the bp | 18:39 | |
| hub_cap | PATCH ftw! | 18:39 |
| dougshelley66 | so question | 18:40 |
| dougshelley66 | on the spec | 18:40 |
| dougshelley66 | is the intention that the key is generated on create? | 18:40 |
| juice | imsplitbit: does the metadata provide any direct functional impact | 18:40 |
| juice | 'or for now is it just metadata? | 18:40 |
| imsplitbit | just metadata atm | 18:40 |
| amrith | implisitbit: +N for a very clear specification | 18:41 |
| imsplitbit | but there is some use for us internal in testing and such | 18:41 |
| imsplitbit | CI | 18:41 |
| juice | do we have any services that will use this and if so does it then imply there needs to be specific metadata in there | 18:41 |
| imsplitbit | juice: not at the moment | 18:41 |
| cp16net | hub_cap: that was the death of me... :-P | 18:41 |
| cp16net | but i think moving forward it will be better :) | 18:41 |
| vipul | right, besides adding labels, what's the use case | 18:41 |
| denis_makogon | glance will have metadata service soon | 18:41 |
| imsplitbit | cp16net and I were talking about using in CI environments to keep track of ci job instances | 18:41 |
| juice | imsplitbit: ok just asking. thanks and a very thorough bp it is | 18:41 |
| imsplitbit | what created them | 18:41 |
| imsplitbit | etc | 18:41 |
| hub_cap | god we need a single metadata svc for openstack hahahah | 18:41 |
| imsplitbit | thanks | 18:41 |
| cp16net | using it as a place for a description or something on an instance | 18:42 |
| imsplitbit | yeah so the entry point in the instance is very small | 18:42 |
| denis_makogon | hub_cap, it's already in process | 18:42 |
| cp16net | or like what tests actually started the job | 18:42 |
| juice | imsplitbit: can you search across instance | 18:42 |
| amrith | two questions. 1. can metadata be specified on instance create and 2. is any metadata stored with an instance snapshot / image that will be automatically instantiated on launch. | 18:42 |
| imsplitbit | once there is an openstack service there's a very clear path to migrate to it | 18:42 |
| juice | impsplitbit: without specifying the instance_id? e.g. show me instances where metadata Key = Value? | 18:42 |
| imsplitbit | 1. not yet, I didn't want to have too much in the inital pr | 18:42 |
| hub_cap | denis_makogon: got a link? is it a mirantis thing? | 18:43 |
| imsplitbit | 2. I don't know that the instance snapshot would store that but we can find a way to hook it in | 18:43 |
| imsplitbit | juice: no | 18:43 |
| hub_cap | i like #1 for a follow on | 18:43 |
| denis_makogon | hub_cap, i'll link you as soon as i reach skype | 18:43 |
| amrith | implisitbit: thanks | 18:43 |
| imsplitbit | instance id is required ATM | 18:43 |
| hub_cap | denis_makogon: npnp | 18:43 |
| juice | imsplitbit: ok just asking | 18:43 |
| juice | :) | 18:43 |
| imsplitbit | np | 18:44 |
| hub_cap | imsplitbit: i approve this bp, personally. i think we need it | 18:44 |
| vipul | imsplitbit: so this doesn't directly relate to the Replication API anymore correct? | 18:44 |
| imsplitbit | I wanted to keep iteration 1 as small as possible | 18:44 |
| imsplitbit | vipul: no it doesn' | 18:44 |
| hub_cap | i also think we need the create stuff too | 18:44 |
| imsplitbit | t | 18:44 |
| hub_cap | but it shoudl be a follow up | 18:44 |
| imsplitbit | although I'm borrowing heavily from it for topology api | 18:44 |
| hub_cap | demorris: will be happy | 18:44 |
| hub_cap | so any issues w/ this? | 18:46 |
| SlickNik | I like the idea; but I'm still trying to wrap my head around the use cases. | 18:47 |
| hub_cap | k | 18:47 |
| hub_cap | lets give some more time | 18:47 |
| hub_cap | to read | 18:47 |
| vipul | SlickNik: +1 I can't really say i don't like it.. i just want to know how metadata would be useful in vNext | 18:47 |
| dougshelley66 | i had a question regarding the name of the key on create | 18:47 |
| dougshelley66 | it looks like it is generated? | 18:47 |
| vipul | like is there certain metadata that will trigger behavior | 18:47 |
| SlickNik | imsplitbit: Can you explain where / how this would be useful for trove? | 18:48 |
| hub_cap | SlickNik: i can give u a use case that isint related to replication | 18:48 |
| imsplitbit | this is less useful to trove and more useful to users of trove | 18:48 |
| hub_cap | tagging for a UI | 18:48 |
| imsplitbit | exactly | 18:48 |
| imsplitbit | users can make use of it's free form to store data that is critical to them to live with the instance that goes beyond the hostname | 18:49 |
| imsplitbit | 'business_purpose': 'this is our super cool reporting box' | 18:49 |
| imsplitbit | or asset tagging for their internal business tracking | 18:49 |
| imsplitbit | it's for the users more than us | 18:49 |
| hub_cap | same reason that say, nova, has it | 18:49 |
| imsplitbit | but we can use it for things liek CI | 18:49 |
| imsplitbit | like | 18:49 |
| *** michael-yu has joined #openstack-trove | 18:49 | |
| hub_cap | man u type liek me | 18:50 |
| hub_cap | ;) | 18:50 |
| vipul | fair enough... you've convinced me | 18:50 |
| imsplitbit | I'm dealing with a pinched nerve | 18:50 |
| SlickNik | Ah, okay. Now that makes a bit more sense to me. | 18:50 |
| kevinconway | so quick quesiton. most of those cases seem like use cases for deployers of trove not developers consuming trove | 18:50 |
| kevinconway | or did i catch that wrong? | 18:50 |
| imsplitbit | our UI team has been requesting it for some time | 18:50 |
| kevinconway | ok not really question. | 18:50 |
| hub_cap | im pretty sure thats the opposite of what we said | 18:50 |
| hub_cap | 18:49 < imsplitbit> 'business_purpose': 'this is our super cool reporting box' | 18:51 |
| hub_cap | how is that deployers? | 18:51 |
| hub_cap | i bought an instance, i need to add biz_purpose to it | 18:51 |
| hub_cap | rax doesnt care | 18:51 |
| abramley | If Nova already has metadata - and Trove deploys on Nova - why couldn't you just put your metadata in the nova instance? How / when do you choose to use Trove metadata vs Nova metadata ? | 18:51 |
| kevinconway | so asset tagging, for example, would need to be enforced by a deployer other wise its just a freeform api call | 18:51 |
| kevinconway | so any developer consuming could NOT add the asset tag | 18:51 |
| imsplitbit | because nova could possibly be obscured from the users depending on how it's deployed | 18:52 |
| hub_cap | if your biz requires asset tagging kevinconway , then you can enforce that w/ your instances you buy | 18:52 |
| juice | Like in my Bp :) | 18:52 |
| hub_cap | but deployers dont care | 18:52 |
| hub_cap | cuz its freeform | 18:52 |
| kevinconway | yeah my brain wires got crossed | 18:52 |
| kevinconway | app devs would use the feature | 18:52 |
| hub_cap | yuppers | 18:52 |
| imsplitbit | yes | 18:52 |
| kevinconway | glad i could help | 18:52 |
| vipul | a trove 'instance' could also be >1 nova instance | 18:53 |
| kevinconway | thank you for your time | 18:53 |
| hub_cap | they have an operation for that.. have u seen pi? | 18:53 |
| imsplitbit | vipul: +1 | 18:53 |
| vgnbkr | Also, if the feature was to be used both by devs and users, wouldn't there need to be some mechanism to determine if the k/v pairs should be shown to users? | 18:54 |
| imsplitbit | vgnbkr: I've put some thought into that | 18:54 |
| hub_cap | vgnbkr: and not deleted ;) | 18:54 |
| imsplitbit | nova solves this by having 3 metadata services | 18:54 |
| imsplitbit | 1 for users | 18:54 |
| hub_cap | LOL srsly?! | 18:54 |
| imsplitbit | 1 for instance internal | 18:54 |
| imsplitbit | and 1 called system_metadata for drivers | 18:54 |
| imsplitbit | totally serious | 18:54 |
| *** jcru has quit IRC | 18:55 | |
| imsplitbit | I don't feel like we need to take it to that leve | 18:55 |
| imsplitbit | l | 18:55 |
| imsplitbit | I thought about adding a .protected method for internal stuff to use and anything protected would not be displayed | 18:55 |
| imsplitbit | but we then take the risk of taking up namespace for keys | 18:55 |
| vgnbkr | That's fine, but then we should clarify that it is only for end user use, no? | 18:55 |
| imsplitbit | this services use is for end user ATM yes | 18:55 |
| SlickNik | So for v1, we (trove) are just going to have 1 right? | 18:56 |
| imsplitbit | yes | 18:56 |
| SlickNik | For the tagging use case for eg, you would need just 1 visible to the end user. | 18:56 |
| SlickNik | Okay, I'm good with it. Nice work on the bp imsplitbit. | 18:56 |
| imsplitbit | thanks! | 18:56 |
| imsplitbit | worked hard on it over the weekend | 18:56 |
| hub_cap | horray, so we all in agreement? | 18:56 |
| vipul | we're not really building a metadata service really.. just a metadata store | 18:56 |
| kevinconway | imsplitbit: can we get a spanish translation of the BP? | 18:56 |
| imsplitbit | correct | 18:56 |
| imsplitbit | metadata store | 18:56 |
| imsplitbit | lol | 18:56 |
| imsplitbit | kevinconway: ^^^ | 18:56 |
| imsplitbit | I had to try to 1 up esp | 18:57 |
| imsplitbit | without doing graphics :) | 18:57 |
| hub_cap | kevinconway: feel free to put up the translations on transifex | 18:57 |
| * imsplitbit is no good with visio | 18:57 | |
| SlickNik | vipul: agreed. I don't think we need to broadcast this to newly booted instances like a metadata "service". | 18:57 |
| hub_cap | oh and kevinconway u had to add a troll remark eh? ;) | 18:57 |
| hub_cap | TROLLLLLLL | 18:57 |
| hub_cap | TROVETROLLLLLL | 18:57 |
| hub_cap | someone bring hiim his crown | 18:57 |
| imsplitbit | sky = blue, kevinconway = trolllllllllllllllllllllll | 18:57 |
| vipul | this was great imsplitbit good job | 18:58 |
| imsplitbit | these are just constants in life | 18:58 |
| hub_cap | imsplitbit: ++ | 18:58 |
| imsplitbit | thanks vipul | 18:58 |
| *** jcru has joined #openstack-trove | 18:58 | |
| hub_cap | kevinconway: hows yer bridgehouse treating ya? | 18:58 |
| hub_cap | TROVETROLLLLLLLLLLL | 18:58 |
| hub_cap | ok so we are done i thikn | 18:58 |
| imsplitbit | I'm working on fixing up some of the feedback from denis and I'll resubmit | 18:59 |
| SlickNik | That's it then, I think we're through with the bp's we had. | 18:59 |
| hub_cap | we def need to clean up the rest of the bps | 18:59 |
| imsplitbit | there is 100% unittest coverage tho | 18:59 |
| hub_cap | annashen_: denis_makogon we will get to your snext wk when they are cleaned up | 18:59 |
| esp | imsplitbit: nice work! | 18:59 |
| denis_makogon | hub_cap, cool | 18:59 |
| SlickNik | Yes, denis_makogon and annashen_ please re-add your bps to the list for the next meeting once they're cleaned up. Thanks! | 18:59 |
| *** kevinconway_ has joined #openstack-trove | 18:59 | |
| denis_makogon | there're still two BPs | 19:00 |
| hub_cap | denis_makogon: none of your sare in the proper format | 19:00 |
| hub_cap | format odes not just mean putting the link | 19:00 |
| hub_cap | plz view the other blueprints | 19:00 |
| hub_cap | and make yours like theirs | 19:00 |
| denis_makogon | point in time recovery and volume data is in proper format | 19:00 |
| SlickNik | denis_makogon: Here is the template: https://wiki.openstack.org/wiki/TroveBlueprint | 19:01 |
| *** kevinconway has quit IRC | 19:01 | |
| *** kevinconway_ is now known as kevinconway | 19:01 | |
| hub_cap | no its not, the blueprint says "point in time recovery" | 19:01 |
| hub_cap | and thats it | 19:01 |
| hub_cap | thats not valid | 19:01 |
| hub_cap | and your other blueprint has a ton of junk in it | 19:01 |
| hub_cap | also | 19:01 |
| hub_cap | not valid | 19:01 |
| grapex | I've got a meeting. Talk to you all later! | 19:02 |
| denis_makogon | fine | 19:02 |
| SlickNik | later grapex | 19:02 |
| denis_makogon | i need some clarification about my BPs now | 19:03 |
| denis_makogon | hub_cap, what do you call "junk" ? | 19:04 |
| *** robertmy_ has quit IRC | 19:05 | |
| *** robertmyers has joined #openstack-trove | 19:05 | |
| denis_makogon | hub_cap, SlickNik, please show me the difference between BP template and https://blueprints.launchpad.net/trove/+spec/point-in-time-recovery, https://blueprints.launchpad.net/trove/+spec/volume-snapshot | 19:05 |
| *** eguz has joined #openstack-trove | 19:06 | |
| denis_makogon | waiting for the response from cores | 19:10 |
| *** eghobo has quit IRC | 19:10 | |
| SlickNik | hi denis_makogon: If you take a look at https://wiki.openstack.org/wiki/Trove/PointInTimeRecovery, it doesn't follow Blueprint template guidelines. I don't see a Justification or impacts section for example. | 19:12 |
| *** demorris has quit IRC | 19:14 | |
| denis_makogon | SlickNik, impacts should be mentioned even if there're none of them ? | 19:14 |
| *** Barker has quit IRC | 19:14 | |
| hub_cap | thank you for handling SlickNik | 19:15 |
| SlickNik | denis_makogon: If a bp has no impact, why chose to do it? :) | 19:15 |
| denis_makogon | SlickNik, for me "impact" mean that this thing ruins/breaks something | 19:16 |
| denis_makogon | SlickNik, and the features are chosen because of the actual use cases and approaches | 19:17 |
| SlickNik | denis_makogon: Ah that's probably a language issue then. "Impact" means what does it affect in *any* way. Not just break / ruin. | 19:17 |
| *** mattgriffin has quit IRC | 19:18 | |
| denis_makogon | SlickNik, and now explain, why did this one skipped https://wiki.openstack.org/wiki/Trove/volume-data-snapshot-design ? It has all sections | 19:21 |
| SlickNik | denis_makogon: BP's are chosen for a wide variety of reasons, most of which are enumerated in the Blueprint template. Which is one of the main reasons why we're asked to submit bps that follow the template guidelines. | 19:22 |
| *** konetzed has joined #openstack-trove | 19:22 | |
| denis_makogon | SlickNik, i get this, but i didn't get the answer why the third BP was skipped | 19:23 |
| denis_makogon | SlickNik, it has __all__ sections from the template | 19:24 |
| denis_makogon | SlickNik, please explain | 19:24 |
| hub_cap | denis_makogon: plz link it | 19:25 |
| hub_cap | id like to see the blueprint, cuz last i checked | 19:25 |
| hub_cap | it had too much junk in the blueprint | 19:25 |
| hub_cap | instead of a nice little pararaph explaining its intent | 19:25 |
| *** mattgriffin has joined #openstack-trove | 19:27 | |
| *** mattgriffin has quit IRC | 19:28 | |
| hub_cap | we want 3 to 5 sentence explainign why, thats it | 19:29 |
| hub_cap | in the blueprint | 19:29 |
| SlickNik | denis_makogon: A couple of things that stand out to me are: | 19:29 |
| SlickNik | 1. It's missing the whole point of _why_ this is needed. If you look at the Justification section, it talks about _HOW_ this can be done, which is not a justification of _WHY_ this should be done. | 19:29 |
| SlickNik | 2. There are essention sections that are needed by the template (eg. Impacts) that have a heading but no text under them. | 19:29 |
| hub_cap | ++ | 19:29 |
| SlickNik | Also, I believe that hub_cap is talking about the actual bp on launchpad (not the wiki-page) i.e. https://blueprints.launchpad.net/trove/+spec/volume-snapshot . This should not be a wall of text. It should just be a few lines regarding the justification, and should link to the wiki page which has the rest of the bp outlined using the template guildelines. | 19:32 |
| *** denis_makogon has quit IRC | 19:33 | |
| openstackgerrit | A change was merged to openstack/python-troveclient: Fix database-list via CLI for large # of dbs https://review.openstack.org/82923 | 19:34 |
| *** rramirez has joined #openstack-trove | 19:37 | |
| *** SnowDust has joined #openstack-trove | 19:39 | |
| *** khyati_ has quit IRC | 19:43 | |
| *** mattgriffin has joined #openstack-trove | 19:44 | |
| hub_cap | yuuuup SlickNik | 19:45 |
| *** radez is now known as radez_g0n3 | 19:49 | |
| *** pdmars has quit IRC | 19:51 | |
| *** pdmars has joined #openstack-trove | 19:51 | |
| ViswaV | "god we need a single metadata svc for openstack hahahah" hub_cap: +42 | 19:57 |
| hub_cap | LOL ViswaV | 19:57 |
| hub_cap | amcrn: dude is the rain comin down in teh wasteland? | 19:58 |
| amcrn | not at the moment | 19:58 |
| hub_cap | man.. its started up here | 19:58 |
| hub_cap | like crazy thunder, lightning, stormy | 19:58 |
| *** mattgriffin has quit IRC | 19:59 | |
| amcrn | yeah, it's supposed to hit down here a little bit later today | 20:01 |
| cp16net | hub_cap: amcrn watch out for the mud slides | 20:02 |
| *** mattgriffin has joined #openstack-trove | 20:12 | |
| *** kevinconway_ has joined #openstack-trove | 20:14 | |
| *** kevinconway has quit IRC | 20:15 | |
| *** kevinconway_ is now known as kevinconway | 20:15 | |
| *** mattgriffin has quit IRC | 20:16 | |
| *** ashestakov has quit IRC | 20:18 | |
| *** mattgriffin has joined #openstack-trove | 20:19 | |
| *** rwsu has joined #openstack-trove | 20:22 | |
| *** kevinconway has quit IRC | 20:24 | |
| *** mattgriffin has quit IRC | 20:27 | |
| *** mattgriffin has joined #openstack-trove | 20:27 | |
| *** khyati_ has joined #openstack-trove | 20:28 | |
| *** michael-yu has quit IRC | 20:30 | |
| *** sbfox has quit IRC | 20:30 | |
| openstackgerrit | Andreas Jaeger proposed a change to openstack/trove: Check that all po/pot files are valid https://review.openstack.org/84239 | 20:35 |
| *** eguz has quit IRC | 20:36 | |
| *** michael-yu has joined #openstack-trove | 20:36 | |
| *** sbfox has joined #openstack-trove | 20:42 | |
| *** tanisdl_ has joined #openstack-trove | 20:46 | |
| *** tanisdl has quit IRC | 20:49 | |
| *** tanisdl_ is now known as tanisdl | 20:49 | |
| *** mattgriffin has quit IRC | 20:49 | |
| *** kevinconway has joined #openstack-trove | 20:50 | |
| *** mattgriffin has joined #openstack-trove | 20:55 | |
| openstackgerrit | Andreas Jaeger proposed a change to openstack/trove: Check that all po/pot files are valid https://review.openstack.org/84239 | 20:56 |
| *** mattgriffin has quit IRC | 20:57 | |
| *** harlowja is now known as harlowja_away | 21:00 | |
| *** mattgriffin has joined #openstack-trove | 21:01 | |
| *** eghobo has joined #openstack-trove | 21:03 | |
| *** mattgriffin has quit IRC | 21:03 | |
| *** pdmars has quit IRC | 21:09 | |
| *** harlowja_away is now known as harlowja | 21:19 | |
| *** robertmyers has quit IRC | 21:30 | |
| *** SnowDust has quit IRC | 21:30 | |
| esp | hey, any core folks in the mood to review? :) https://review.openstack.org/#/c/81682/, https://review.openstack.org/#/c/81557/, https://review.openstack.org/#/c/81410/ | 21:31 |
| esp | or non-core ^ | 21:31 |
| *** esp has left #openstack-trove | 21:31 | |
| *** sbfox has quit IRC | 21:33 | |
| *** grapex has quit IRC | 21:35 | |
| *** sbfox has joined #openstack-trove | 21:38 | |
| *** esp has joined #openstack-trove | 21:39 | |
| *** michael-yu has quit IRC | 21:46 | |
| *** ViswaV has quit IRC | 21:51 | |
| *** michael-yu has joined #openstack-trove | 21:52 | |
| *** ViswaV has joined #openstack-trove | 21:53 | |
| *** flaper87 is now known as flaper87|afk | 21:57 | |
| *** michael-yu has quit IRC | 22:12 | |
| *** michael-yu has joined #openstack-trove | 22:16 | |
| *** jasonb365 has joined #openstack-trove | 22:23 | |
| *** kevinconway has quit IRC | 22:26 | |
| *** jcru has quit IRC | 22:27 | |
| *** yogesh has quit IRC | 22:28 | |
| *** jasonb365 has quit IRC | 22:31 | |
| *** amytron has quit IRC | 22:31 | |
| *** esp2 has joined #openstack-trove | 22:40 | |
| *** esp2 has quit IRC | 22:45 | |
| *** grapex has joined #openstack-trove | 23:03 | |
| *** jmontemayor has quit IRC | 23:13 | |
| openstackgerrit | Nikhil Manchanda proposed a change to openstack/trove: Remove mockito, and replace with mock https://review.openstack.org/82888 | 23:23 |
| *** khyati_ has quit IRC | 23:25 | |
| *** yidclare has quit IRC | 23:26 | |
| *** rramirez has quit IRC | 23:32 | |
| *** tanisdl has quit IRC | 23:41 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
