| *** vipul is now known as vipul-away | 00:20 | |
| *** mmcdaris has quit IRC | 00:37 | |
| *** vipul-away is now known as vipul | 00:41 | |
| *** matsuhashi has joined #openstack-trove | 00:54 | |
| *** nosnos has joined #openstack-trove | 01:00 | |
| *** robertmyers has joined #openstack-trove | 01:16 | |
| *** erkules_ has joined #openstack-trove | 01:24 | |
| *** erkules has quit IRC | 01:27 | |
| *** erkules_ has quit IRC | 01:46 | |
| *** erkules_ has joined #openstack-trove | 01:47 | |
| *** fandikurnia01 has joined #openstack-trove | 01:50 | |
| *** fandikurnia01 has quit IRC | 02:07 | |
| *** vipul has quit IRC | 02:12 | |
| *** vipul has joined #openstack-trove | 02:13 | |
| *** vipul is now known as vipul-away | 02:28 | |
| *** robertmyers has quit IRC | 02:51 | |
| *** vipul-away is now known as vipul | 02:53 | |
| *** fandikurnia01 has joined #openstack-trove | 03:05 | |
| *** vipul is now known as vipul-away | 03:09 | |
| *** Kapil has joined #openstack-trove | 04:29 | |
| *** SushilKM__ has joined #openstack-trove | 04:58 | |
| *** adrian_otto has joined #openstack-trove | 05:08 | |
| *** adrian_otto has quit IRC | 05:08 | |
| *** redthrux has quit IRC | 05:21 | |
| *** kevinconway has quit IRC | 05:21 | |
| *** Nate3 has quit IRC | 05:21 | |
| *** Nate1 has joined #openstack-trove | 05:22 | |
| *** kevinconway has joined #openstack-trove | 05:22 | |
| *** redthrux has joined #openstack-trove | 05:24 | |
| *** vipul-away is now known as vipul | 05:36 | |
| *** nosnos has quit IRC | 05:36 | |
| *** nosnos_ has joined #openstack-trove | 05:36 | |
| *** matsuhashi has quit IRC | 05:56 | |
| *** erkules_ is now known as erkules | 06:02 | |
| *** vipul is now known as vipul-away | 06:05 | |
| *** vipul-away is now known as vipul | 06:24 | |
| *** dafter has quit IRC | 06:26 | |
| *** dafter has joined #openstack-trove | 06:26 | |
| *** matsuhashi has joined #openstack-trove | 06:30 | |
| *** dafter has quit IRC | 06:30 | |
| *** vipul is now known as vipul-away | 06:40 | |
| *** mmcdaris has joined #openstack-trove | 06:41 | |
| *** vipul-away is now known as vipul | 06:43 | |
| *** matsuhashi has quit IRC | 06:51 | |
| *** matsuhashi has joined #openstack-trove | 06:54 | |
| *** vipul is now known as vipul-away | 07:00 | |
| *** matsuhas_ has joined #openstack-trove | 07:01 | |
| *** matsuhashi has quit IRC | 07:01 | |
| *** nosnos has joined #openstack-trove | 07:02 | |
| *** nosnos_ has quit IRC | 07:02 | |
| *** vipul-away is now known as vipul | 07:04 | |
| *** nosnos has quit IRC | 07:19 | |
| *** nosnos has joined #openstack-trove | 07:20 | |
| *** vipul is now known as vipul-away | 07:25 | |
| *** vipul-away is now known as vipul | 07:27 | |
| *** vipul is now known as vipul-away | 07:43 | |
| *** vipul-away is now known as vipul | 07:52 | |
| *** mmcdaris has quit IRC | 07:55 | |
| *** SnowDust has joined #openstack-trove | 08:42 | |
| SnowDust | hello all .. ! | 08:42 |
|---|---|---|
| *** ashestakov_ has joined #openstack-trove | 08:46 | |
| *** ashestakov_ has quit IRC | 08:48 | |
| dmakogon | hi | 08:56 |
| SnowDust | hello dmakogon | 09:27 |
| dmakogon | SnowDust: whaazzzzzzzzzzzzaaaaaaaaaaaaaaaap | 09:27 |
| SnowDust | nothing new .. bloated .. | 09:27 |
| SnowDust | share something xciting ! | 09:27 |
| dmakogon | what do you want to see ? | 09:28 |
| dmakogon | do you want me to do knock-knock jokes ? | 09:28 |
| *** ppenjoy has joined #openstack-trove | 09:31 | |
| SnowDust | no .. may be .. ur code .. | 09:32 |
| SnowDust | even that does charm me ! | 09:32 |
| dmakogon | oh, i'm just updating my last reviews | 09:33 |
| dmakogon | to get finally +1 | 09:33 |
| dmakogon | +2 | 09:33 |
| SnowDust | yeah .. some brilliant ones .. need the world to catchup with them .. :) | 09:33 |
| SnowDust | but a +2 on your code quality | 09:34 |
| SnowDust | always ! | 09:34 |
| SnowDust | envy u ! | 09:34 |
| SnowDust | ;) | 09:34 |
| *** ashestakov_ has joined #openstack-trove | 09:35 | |
| dmakogon | so, any news ? | 09:37 |
| dmakogon | how was you weekend ? | 09:37 |
| SnowDust | kind .. of .. wrecked :) | 09:38 |
| SnowDust | am salvaging it .. :) | 09:38 |
| openstackgerrit | Denis M. proposed a change to openstack/trove: Update statuses on GA timeout https://review.openstack.org/45723 | 09:44 |
| *** ppenjoy has left #openstack-trove | 10:32 | |
| *** SushilKM__ has quit IRC | 10:32 | |
| *** fandikurnia01 has quit IRC | 10:59 | |
| *** fandikurnia01 has joined #openstack-trove | 11:00 | |
| *** SushilKM has joined #openstack-trove | 11:00 | |
| *** SushilKM has quit IRC | 11:04 | |
| *** fandikurnia01 has quit IRC | 11:04 | |
| *** fandikurnia01 has joined #openstack-trove | 11:05 | |
| openstackgerrit | Denis M. proposed a change to openstack/trove: PEP8. E125, F811, F401, H703 https://review.openstack.org/48305 | 11:08 |
| *** ikhudoshyn_ has joined #openstack-trove | 11:28 | |
| *** ikhudoshyn has quit IRC | 11:28 | |
| *** dukhlov has quit IRC | 11:37 | |
| *** dukhlov has joined #openstack-trove | 11:39 | |
| *** fandikurnia01 has quit IRC | 11:59 | |
| *** SushilKM has joined #openstack-trove | 11:59 | |
| *** pdmars has joined #openstack-trove | 12:03 | |
| *** matsuhas_ has quit IRC | 12:04 | |
| *** matsuhashi has joined #openstack-trove | 12:06 | |
| *** matsuhashi has quit IRC | 12:10 | |
| *** matsuhashi has joined #openstack-trove | 12:11 | |
| *** Kapil has quit IRC | 12:14 | |
| *** rushiagr has joined #openstack-trove | 12:21 | |
| *** fandikurnia01 has joined #openstack-trove | 12:24 | |
| *** SnowDust has quit IRC | 12:36 | |
| *** nosnos has quit IRC | 12:47 | |
| *** rustlebee is now known as russellb | 12:54 | |
| *** matsuhashi has quit IRC | 12:55 | |
| *** matsuhashi has joined #openstack-trove | 12:56 | |
| *** matsuhashi has quit IRC | 13:00 | |
| *** matsuhashi has joined #openstack-trove | 13:01 | |
| *** jlucci has joined #openstack-trove | 13:07 | |
| *** matsuhashi has quit IRC | 13:13 | |
| *** matsuhashi has joined #openstack-trove | 13:14 | |
| *** matsuhashi has quit IRC | 13:18 | |
| *** robertmyers has joined #openstack-trove | 13:19 | |
| openstackgerrit | Pradeep Kumar Chandani proposed a change to openstack/trove: Service Registration using conf file https://review.openstack.org/41055 | 13:32 |
| *** jmontemayor has joined #openstack-trove | 13:41 | |
| ikhudoshyn_ | PradeepChandani: Hi! There is a consideration about ur review | 13:50 |
| ikhudoshyn_ | Looks like defaults in cfg.py have higher priority than user configured service type registry. Is it intended? | 13:53 |
| *** rushiagr has quit IRC | 13:57 | |
| *** jcru has joined #openstack-trove | 13:59 | |
| *** dafter has joined #openstack-trove | 14:01 | |
| *** Barker has joined #openstack-trove | 14:02 | |
| *** kevinconway has quit IRC | 14:06 | |
| *** dafter has quit IRC | 14:15 | |
| *** dukhlov has quit IRC | 14:37 | |
| *** dukhlov has joined #openstack-trove | 14:37 | |
| openstackgerrit | Pradeep Kumar Chandani proposed a change to openstack/trove: Service Registration using conf file https://review.openstack.org/41055 | 14:39 |
| *** kevinconway has joined #openstack-trove | 14:46 | |
| *** Kapil has joined #openstack-trove | 14:49 | |
| *** amytron has joined #openstack-trove | 15:03 | |
| *** SushilKM has quit IRC | 15:03 | |
| *** rnirmal has joined #openstack-trove | 15:06 | |
| *** tanisdl has joined #openstack-trove | 15:09 | |
| *** adrian_otto has joined #openstack-trove | 15:10 | |
| *** adrian_otto has quit IRC | 15:18 | |
| *** paul_lodronio has joined #openstack-trove | 15:20 | |
| *** yidclare has joined #openstack-trove | 15:25 | |
| *** Barker has quit IRC | 15:30 | |
| *** Barker has joined #openstack-trove | 15:34 | |
| *** erkules_ has joined #openstack-trove | 15:37 | |
| *** Kapil has quit IRC | 15:39 | |
| *** erkules has quit IRC | 15:39 | |
| *** Nate1 has quit IRC | 16:02 | |
| dmakogon | anybody home ? | 16:02 |
| *** Nate1 has joined #openstack-trove | 16:05 | |
| *** adrian_otto has joined #openstack-trove | 16:14 | |
| *** dukhlov has quit IRC | 16:19 | |
| *** dukhlov has joined #openstack-trove | 16:20 | |
| *** rushiagr has joined #openstack-trove | 16:32 | |
| *** adrian_otto has quit IRC | 16:34 | |
| *** mmcdaris has joined #openstack-trove | 16:41 | |
| *** mmcdaris has quit IRC | 16:46 | |
| *** mmcdaris has joined #openstack-trove | 16:48 | |
| *** mmcdaris has quit IRC | 16:52 | |
| *** Nate1 has quit IRC | 16:54 | |
| *** mmcdaris has joined #openstack-trove | 16:56 | |
| *** adrian_otto has joined #openstack-trove | 16:57 | |
| esmute | Hi dmakogon | 16:58 |
| hub_cap | lol dmakogon | 16:58 |
| dmakogon | hi | 16:58 |
| dmakogon | my nick is normal | 16:59 |
| dmakogon | until i connect from home))) | 16:59 |
| hub_cap | ikhudoshyn_: thx for catching that, user configured should override it PradeepChandani | 16:59 |
| hub_cap | dmakogon: set up a bouncer or irssi proxy!! ;) | 16:59 |
| hub_cap | im on my irssi server right now chatting with you but behind my terminal window i have my irc client, logged in as me too (to the irssi proxy) | 17:00 |
| kevinconway | i connect to irc using TOR via my wristwatch | 17:01 |
| kevinconway | and type messages by making funny facial gestures | 17:01 |
| isvirido` | kevinconway, i would like to see it | 17:02 |
| *** SnowDust has joined #openstack-trove | 17:02 | |
| *** yidclare has quit IRC | 17:02 | |
| hub_cap | kevinconway: omg lol | 17:04 |
| hub_cap | isvirido`: by wristwatch he means http://www.jeremyblum.com/wp-content/uploads/2013/05/2013-05-26-09.15.45-e1369896493547.jpg | 17:05 |
| isvirido` | poor, kevinconway ... | 17:07 |
| *** SushilKM has joined #openstack-trove | 17:08 | |
| hub_cap | :P | 17:10 |
| openstackgerrit | Denis M. proposed a change to openstack/trove: Security groups workflow update https://review.openstack.org/45708 | 17:12 |
| SnowDust | isvirido: have u tried shell scripting from heat ? | 17:12 |
| *** rushiagr has quit IRC | 17:13 | |
| dmakogon | hub_cap: https://review.openstack.org/#/c/48305/ | 17:13 |
| dmakogon | hub_cap: https://review.openstack.org/#/c/45723/ | 17:13 |
| dmakogon | hub_cap: just as i promise | 17:13 |
| hub_cap | :) | 17:13 |
| *** tvb has joined #openstack-trove | 17:14 | |
| *** SushilKM has quit IRC | 17:15 | |
| *** SushilKM__ has joined #openstack-trove | 17:15 | |
| *** tvb is now known as dafter | 17:15 | |
| openstackgerrit | Denis M. proposed a change to openstack/trove: Security groups workflow update https://review.openstack.org/45708 | 17:15 |
| SnowDust | isvirido`: gone ????? | 17:16 |
| dmakogon | forgot to ask tox if my code is beautiful | 17:16 |
| dmakogon | SnowDust: a bit | 17:16 |
| *** mmcdaris has quit IRC | 17:16 | |
| SnowDust | isvirido` do` do` da` di` du` de` di` | 17:19 |
| vipul | dmakogon: https://review.openstack.org/#/c/45723/14/trove/common/instance.py | 17:20 |
| dmakogon | vipul: look at patch 21 | 17:21 |
| dmakogon | vipul: or i'm missing something ? | 17:23 |
| hub_cap | dmakogon: what do u use to code? ide ? | 17:25 |
| vipul | dmakogon: Nope, I just question intorducing a new status | 17:25 |
| vipul | dmakogon: I am not sure if we've already discussed this, if so, let me know | 17:26 |
| dmakogon | hub_cap: PyCharm | 17:26 |
| hub_cap | PyCharm should have flake8 integration, id think | 17:26 |
| dmakogon | vipul: i think this status is needed | 17:26 |
| dmakogon | vipul: and yes, we've discussed it already | 17:27 |
| dmakogon | hub_cap: yes, but not restricted, or sometimes i'm missing it | 17:27 |
| dmakogon | hub_cap: sorry for flooding | 17:27 |
| hub_cap | ah dmakogon, no worries. u will learn to run tox before u push :) | 17:29 |
| *** harlowja has joined #openstack-trove | 17:29 | |
| dmakogon | hub_cap: i already know that, but i forgot it, tired =( | 17:30 |
| hub_cap | :P | 17:30 |
| hub_cap | dmakogon: u can set up post commit hooks too | 17:30 |
| hub_cap | that just run it for u | 17:30 |
| hub_cap | then no matter how tired u are! | 17:30 |
| hub_cap | itll run!! | 17:30 |
| *** mmcdaris has joined #openstack-trove | 17:31 | |
| *** yogesh has joined #openstack-trove | 17:32 | |
| dmakogon | hub_cap: ok | 17:33 |
| hub_cap | just a thought... i havent done it but hey, it might be fun to do | 17:34 |
| *** arborism has joined #openstack-trove | 17:35 | |
| dmakogon | hub_cap: what do you mean ? | 17:36 |
| hub_cap | it might be fun to add post commit hooks | 17:38 |
| hub_cap | to run tox | 17:38 |
| hub_cap | then blog about it so others can do it :) | 17:38 |
| dmakogon | only tox -epep8 | 17:38 |
| hub_cap | ya | 17:39 |
| dmakogon | but idea is really cool | 17:39 |
| *** arborism is now known as amrcn | 17:39 | |
| *** amrcn is now known as amcrn | 17:39 | |
| dmakogon | guys, i have and idea for jenkins random falls | 17:42 |
| *** dafter has quit IRC | 17:42 | |
| dmakogon | what about using soft asserts ? | 17:42 |
| *** dafter has joined #openstack-trove | 17:42 | |
| *** dafter has quit IRC | 17:42 | |
| *** dafter has joined #openstack-trove | 17:42 | |
| dmakogon | suppose we have assert AssertionError: 2013-10-07T17:41:09Z != 2013-10-07T17:41:08Z, what if we would use something like assert.AlmostEquals ? | 17:45 |
| hub_cap | if we have time bugs dmakogon, those are bugs | 17:45 |
| hub_cap | thats definitely a bug, weve fixed like ~3 of those already | 17:46 |
| *** dafter has quit IRC | 17:46 | |
| *** dafter has joined #openstack-trove | 17:48 | |
| *** dafter has quit IRC | 17:48 | |
| *** dafter has joined #openstack-trove | 17:48 | |
| *** mmcdaris_ has joined #openstack-trove | 17:49 | |
| *** mmcdaris has quit IRC | 17:49 | |
| *** mmcdaris_ is now known as mmcdaris | 17:49 | |
| *** tvb|afk has joined #openstack-trove | 17:49 | |
| *** tvb|afk has quit IRC | 17:49 | |
| *** tvb|afk has joined #openstack-trove | 17:49 | |
| *** adrian_otto has quit IRC | 17:51 | |
| *** dafter has quit IRC | 17:53 | |
| openstackgerrit | Denis M. proposed a change to openstack/trove: Security groups workflow update https://review.openstack.org/45708 | 17:53 |
| *** tvb|afk has quit IRC | 17:53 | |
| *** yidclare has joined #openstack-trove | 17:54 | |
| *** dafter has joined #openstack-trove | 17:55 | |
| *** tvb|afk has joined #openstack-trove | 17:59 | |
| *** dafter has quit IRC | 18:01 | |
| *** jlucci has quit IRC | 18:06 | |
| *** dafter has joined #openstack-trove | 18:09 | |
| *** tvb|afk has quit IRC | 18:12 | |
| dmakogon | vipul: https://review.openstack.org/#/c/45708/ | 18:16 |
| vipul | SlickNik: https://review.openstack.org/#/c/45238/ Please review.. The int-tests seem to fail on Queue Delete.. is there something missed in this patch? | 18:17 |
| openstackgerrit | A change was merged to openstack/trove: PEP8. E125, F811, F401, H703 https://review.openstack.org/48305 | 18:20 |
| juice | hub_cap: grapex: datsun180b: I know a lot is on your plate however, this review is a fix for a "critical" bug for us. It impacts our billing process. https://review.openstack.org/#/c/49855/ | 18:23 |
| SlickNik | vipul / dukhlov: Yes, will take a look. On first glance, there definitely seems to be an issue with the patch, and not with the tests. | 18:23 |
| hub_cap | juice: cry me a river | 18:24 |
| hub_cap | JKJKJKJK | 18:24 |
| SlickNik | juice: you said "done" to one of the comments. Are you planning to upload a new patch-set for that? | 18:25 |
| hub_cap | hahah | 18:25 |
| hub_cap | juice: u do win for the ugliest import ever | 18:26 |
| hub_cap | from trove.tests.api.instances import GROUP as INSTANCE_GROUP, instance_info | 18:26 |
| hub_cap | fix that shit | 18:26 |
| juice | I was just following by example :) | 18:27 |
| juice | I hit git review but didn't notice that it came back with a rebase error | 18:27 |
| hub_cap | HAHA | 18:27 |
| hub_cap | oopsies | 18:27 |
| hub_cap | good | 18:27 |
| hub_cap | then it wouldnt have merged anyway | 18:27 |
| juice | btw - can we discuss our "import" statement styles in the near future | 18:27 |
| hub_cap | so yer getting 2 birds stoned at once | 18:27 |
| SlickNik | heh | 18:27 |
| hub_cap | yes juice for sure | 18:28 |
| hub_cap | we can name it the juice rule | 18:28 |
| *** Nate1 has joined #openstack-trove | 18:29 | |
| dukhlov | Hi! | 18:30 |
| juice | hub_cap: classic | 18:30 |
| hub_cap | :P | 18:30 |
| hub_cap | dukhlov: hello | 18:30 |
| *** demorris has joined #openstack-trove | 18:30 | |
| hub_cap | raiders game went till midnight last night | 18:31 |
| hub_cap | :o | 18:31 |
| *** grapex has joined #openstack-trove | 18:31 | |
| dukhlov | regarding this | 18:31 |
| dukhlov | https://review.openstack.org/#/c/45238/ | 18:31 |
| dukhlov | I've tested on my local VM | 18:32 |
| openstackgerrit | Justin Hopper proposed a change to openstack/trove: Fix User ID in Context and Notifications https://review.openstack.org/49855 | 18:32 |
| amcrn | hub_cap: they beat my chargers :/ | 18:32 |
| dukhlov | and got the same problem | 18:32 |
| *** SnowDust has quit IRC | 18:32 | |
| juice | ok - the import statement is fixed | 18:32 |
| dukhlov | so now I'm trying to find out problem in code | 18:32 |
| hub_cap | amcrn: :) | 18:33 |
| hub_cap | we were on the bart yest going to bluegrass festival and i met a texan...... i could tell cuz he was teh only rider going to candlestick w/o a 9ers shirt on.. he was in full texans garb | 18:34 |
| *** SnowDust has joined #openstack-trove | 18:34 | |
| hub_cap | so one might say i made a dumb decision | 18:35 |
| hub_cap | i fly back from hong kong and the next day im flying to vegas for my wifes bday | 18:35 |
| juice | not really your jet lag should play nicely with the vegas night life | 18:36 |
| juice | sleep all day - play all night | 18:36 |
| juice | though I just had the image of you drooling in a half-comatose state during your nice romantic steak dinner and you wife glaring at you with this "Really?!" expression on her face | 18:37 |
| hub_cap | lol steak.. hardly | 18:37 |
| hub_cap | more like raw bar | 18:37 |
| hub_cap | last bday was king crab and martinis :) | 18:38 |
| hub_cap | in vegas that is... | 18:38 |
| amcrn | No Joel Robuchon? Pft. | 18:38 |
| hub_cap | let me try to find | 18:38 |
| hub_cap | it was some chef place | 18:39 |
| hub_cap | but we just got king crap and pomegranite martinis and some other raw bar stuff | 18:39 |
| hub_cap | *crab lol | 18:40 |
| hub_cap | twas still great | 18:40 |
| hub_cap | i cant believe i can get to/from vegas for < 200/pp .. its crazy amytron | 18:43 |
| hub_cap | amcrn: ^ ^ | 18:43 |
| hub_cap | lol @ pinging amytron TAB FAIL | 18:43 |
| openstackgerrit | A change was merged to openstack/trove: Service Registration using conf file https://review.openstack.org/41055 | 18:44 |
| amytron | well. for what's it worth, i can't believe that either, hub_cap | 18:44 |
| amytron | what it's* | 18:45 |
| hub_cap | right? | 18:45 |
| amcrn | can someone co-sign on a concept real quick? I want to do image building for trove on a completely different box, but I find that keystone/glance are hardcoded to localhost. I'll submit a patch (so it can be assigned via an rc), but I want to first confirm my intentions aren't bizarre. | 18:46 |
| hub_cap | grapex: around? | 18:46 |
| grapex | hub_cap: Yes. | 18:46 |
| grapex | I just got back on. | 18:46 |
| grapex | On a loaner Mac | 18:46 |
| amcrn | in general, I don't want to have to build images on the same box the api/service is running on, I don't see why the coupling is necessary. | 18:46 |
| hub_cap | so youre saying that dib depends on them? | 18:47 |
| hub_cap | or our redstack code depends on them being localhost? | 18:47 |
| amcrn | latter | 18:47 |
| hub_cap | so here the deal from my perspective | 18:48 |
| hub_cap | and im w/ you ripping it apart | 18:48 |
| hub_cap | imho, redstack is for _local_ dev only | 18:48 |
| hub_cap | as in, dev/testing | 18:48 |
| hub_cap | building a multi node env or a actualy env for running this should be left to things that arent a crappy bash script | 18:48 |
| hub_cap | so im cool w/ you changing it to optionally be off the node youre building | 18:49 |
| hub_cap | but i caution you to not rely on redstack lol | 18:49 |
| amcrn | You read my mind anyway, I was asking from the vantage point of the script, but I'm puppet'izing it ;) | 18:49 |
| hub_cap | :) | 18:50 |
| hub_cap | oh and dude puppet is so last wk | 18:50 |
| hub_cap | its salt now ;) | 18:50 |
| amcrn | you're telling me | 18:50 |
| amcrn | puppet and i are far from friends, let me leave it at that | 18:51 |
| *** vipul is now known as vipul-away | 18:53 | |
| hub_cap | lol amcrn | 18:53 |
| *** grapex has quit IRC | 18:53 | |
| *** grapex has joined #openstack-trove | 18:54 | |
| *** SnowDust has quit IRC | 18:54 | |
| *** amytron has quit IRC | 18:55 | |
| *** amytron has joined #openstack-trove | 18:55 | |
| *** SnowDust has joined #openstack-trove | 18:56 | |
| *** mmcdaris has quit IRC | 19:02 | |
| *** vipul-away is now known as vipul | 19:04 | |
| dukhlov | As I understood now trove is a part of devstack. Is it correct? | 19:07 |
| dukhlov | And problem with integration tests connected to this code migration | 19:08 |
| dukhlov | ? | 19:08 |
| hub_cap | no dukhlov | 19:08 |
| hub_cap | we made sure of no issues | 19:08 |
| *** SushilKM__ has quit IRC | 19:08 | |
| hub_cap | https://rdjenkins.dyndns.org/job/Trove-Gate/ <-- see the green | 19:08 |
| hub_cap | there was an issue last wk that we fixed in redstack though | 19:09 |
| hub_cap | https://github.com/openstack/trove-integration/commit/82df245073429cf5b4747841494f46b75119f2ec | 19:09 |
| openstackgerrit | Justin Hopper proposed a change to openstack/trove: Fix User ID in Context and Notifications https://review.openstack.org/49855 | 19:11 |
| *** shivamshukla_ has joined #openstack-trove | 19:17 | |
| *** dafter has quit IRC | 19:18 | |
| juice | dmakogon: I have a fix for the intermittent timestamp error in tests | 19:20 |
| juice | putting together a patch now | 19:20 |
| *** amcrn has quit IRC | 19:21 | |
| *** adrian_otto has joined #openstack-trove | 19:26 | |
| openstackgerrit | Michael Basnight proposed a change to openstack/python-troveclient: Massive refactoring to the troveclient. - WIP until further notice. https://review.openstack.org/48576 | 19:35 |
| *** dafter has joined #openstack-trove | 19:35 | |
| *** dafter has quit IRC | 19:35 | |
| *** dafter has joined #openstack-trove | 19:35 | |
| *** mmcdaris has joined #openstack-trove | 19:35 | |
| hub_cap | kevinconway: around? | 19:37 |
| kevinconway | hub_cap: yeah, i'm here. whats up? | 19:39 |
| hub_cap | was wondering if u could see tim... i know where he is now (or isint lol) | 19:40 |
| *** mmcdaris has quit IRC | 19:40 | |
| *** Nate1 has quit IRC | 19:45 | |
| *** tanisdl has quit IRC | 19:46 | |
| *** mmcdaris has joined #openstack-trove | 19:49 | |
| *** tanisdl has joined #openstack-trove | 19:49 | |
| *** Nate1 has joined #openstack-trove | 19:50 | |
| *** Nate1 has quit IRC | 19:50 | |
| *** Nate1 has joined #openstack-trove | 19:50 | |
| *** dmakogon_ipod has joined #openstack-trove | 19:50 | |
| *** yogesh has quit IRC | 19:51 | |
| *** dukhlov has quit IRC | 19:56 | |
| SlickNik | amcrn: have you tried setting the SERVICE_HOST parameter? | 19:57 |
| dmakogon_ipod | amcrn: i have posted few comment on statuses update review. | 20:00 |
| *** yogesh has joined #openstack-trove | 20:00 | |
| dmakogon_ipod | yogesh: hi | 20:01 |
| dmakogon_ipod | what's you nick at launchpad ? | 20:01 |
| dmakogon_ipod | yogesh:i'd like to reassign my BP on you | 20:01 |
| yogesh | yogesh-mehra | 20:02 |
| yogesh | hey... | 20:02 |
| imsplitbit | vipul: yt? | 20:03 |
| dmakogon_ipod | yogesh: done | 20:03 |
| imsplitbit | vipul: ICMP ECHO | 20:03 |
| imsplitbit | vipul: SYN | 20:03 |
| imsplitbit | vipul: SYN | 20:03 |
| imsplitbit | :) | 20:03 |
| hub_cap | your packets are dropping | 20:03 |
| imsplitbit | I can see that | 20:03 |
| hub_cap | watchout vipul imsplitbit is trying to ddos you | 20:04 |
| vipul | imsplitbit: ELHO | 20:04 |
| imsplitbit | sweet | 20:04 |
| vipul | what's up | 20:04 |
| imsplitbit | ok lets talk replication | 20:04 |
| imsplitbit | https://etherpad.openstack.org/TroveClusteringAPI-WIP | 20:05 |
| imsplitbit | hub_cap and I have been discussing the replication capability of a service type | 20:05 |
| imsplitbit | and how to represent it | 20:05 |
| hub_cap | oh dont bring me into this | 20:05 |
| imsplitbit | lol | 20:05 |
| imsplitbit | you weren't discussing it with me? | 20:05 |
| imsplitbit | did I completely imagine this? | 20:05 |
| imsplitbit | SHUT IT! | 20:06 |
| vipul | has that page been updated since last? | 20:06 |
| hub_cap | lol imsplitbit | 20:06 |
| imsplitbit | no but just in case you needed a refresher on what you and hub_cap talked about | 20:06 |
| imsplitbit | I just linked it for context | 20:06 |
| imsplitbit | and for others wanting to participate | 20:06 |
| dmakogon_ipod | hub_cap: hi, again | 20:06 |
| hub_cap | hi dmakogon_ipod | 20:07 |
| imsplitbit | so the conclusion drawn was that while replication is a *capability* of a service type, it would be represented just as a cluster is represented | 20:07 |
| imsplitbit | and I'm not in agreement with that statement | 20:07 |
| imsplitbit | so I would love to discuss with you | 20:07 |
| imsplitbit | my contention is that for anything other than simple master/slave relationship this representation of the data is not clear. | 20:08 |
| imsplitbit | i.e. using "nodes" to represent the instances that are a part of the cluster | 20:08 |
| imsplitbit | for example: if you want to make a slave of a slave | 20:09 |
| imsplitbit | this then makes the metadata structure a hot mess | 20:09 |
| imsplitbit | or rather lets say you want to have 1 master with 5 slaves and each of those 5 slaves has a few slaves | 20:09 |
| dmakogon_ipod | i'd like you, all, to take a look at this review https://review.openstack.org/#/c/45708/ | 20:09 |
| dmakogon_ipod | imsplitbit: mongo db with sharding is a good example | 20:10 |
| vipul | Sure.. if you are implying a hiearchy that's n levels deep.. then it will break down | 20:10 |
| imsplitbit | representing that data in a nested object within the metatdata of the instance object doesn't seem to be the cleanest way to represent it | 20:10 |
| imsplitbit | right | 20:10 |
| vipul | But say you are doing a master-master-master setup | 20:10 |
| vipul | how is that represented? why can't it be represented as a cluster | 20:11 |
| imsplitbit | I was proposing to keep instnaces that are replicated separate with metadata representing what their relationship is to one another in their individual metadata | 20:11 |
| vipul | Yea so keep it one level | 20:11 |
| imsplitbit | what if you want a read only slave from that master*3 setup? for reporting | 20:11 |
| imsplitbit | right | 20:11 |
| imsplitbit | each master would be it's own instnace | 20:11 |
| imsplitbit | with metadata that says "hey, I'm a slave of another instance and here is it's ref | 20:12 |
| vipul | But if you only do multi-master.. than you can have one instance | 20:12 |
| imsplitbit | I wouldn't disagree with that except that then you have an inconsistency in how the data is represented | 20:12 |
| imsplitbit | where are the rules that define what is what? | 20:13 |
| imsplitbit | if masters > 1 then represent it as a single instance? | 20:13 |
| vipul | So how about this.. why can't nodes be elevated to instances or consider them interchangeable | 20:13 |
| vipul | In the event you add a new slave node | 20:14 |
| vipul | you can referece the Master as a Node instead of an instance | 20:14 |
| imsplitbit | well but then what's the value of adding nodes to the api? | 20:14 |
| imsplitbit | if they're just instances? | 20:14 |
| vipul | They are represented as 'nodes' still.. | 20:14 |
| vipul | just that there are scnearios that allow you to group something to other Instances.. or in the case of replication.. other nodes | 20:15 |
| imsplitbit | but where is a clean line of separation between when something gets displayed as an instance with node subresources and when the instances are displayed separately? | 20:15 |
| vipul | I think we can say that we always have node subresources | 20:16 |
| vipul | An instance just becomes a container that groups nodes | 20:16 |
| imsplitbit | so every instance would have at least 1 node that is itself? | 20:16 |
| *** isvirido` has quit IRC | 20:16 | |
| vipul | yes | 20:16 |
| *** SnowDust has quit IRC | 20:16 | |
| vipul | there is no Physical instance.. an instance is just a collection maybe of one | 20:16 |
| * imsplitbit is pondering heavily | 20:16 | |
| imsplitbit | let me sketch this out | 20:17 |
| vipul | That allows you to configure at the Node level.. which then makes the replication use case the same as cluster | 20:17 |
| imsplitbit | I'm doing this on the etherpad if you want to follow/edit | 20:17 |
| vipul | Sure.. | 20:17 |
| vipul | The only differnce is with Repication.. you may have depth of N | 20:18 |
| vipul | which likely you wouldn't have with a cluster | 20:18 |
| vipul | Actually you could if you wanted to.. like set up slaves for every Galera node | 20:18 |
| hub_cap | right but u can rep that as replication | 20:18 |
| hub_cap | either w separate instances or not | 20:18 |
| vipul | sure.. SO you have 1 instance representing a 3 node cluster.. and 3 other separate instances that are slaves.. which replicate from a 'node' in a cluster' | 20:21 |
| imsplitbit | see I would see that as different instances | 20:21 |
| imsplitbit | 1 representing the clsuter | 20:21 |
| imsplitbit | cluster | 20:21 |
| imsplitbit | damnit I can't type today | 20:21 |
| imsplitbit | and each slave is it's own instance because it isn't necessarily part of that cluster | 20:22 |
| hub_cap | u suck imsplitbit | 20:22 |
| imsplitbit | it's just a replicated copy of that clustered data | 20:22 |
| hub_cap | ;) | 20:22 |
| imsplitbit | lol | 20:22 |
| imsplitbit | I know | 20:22 |
| imsplitbit | my dad used to tell me that all growing up while he beat me | 20:22 |
| imsplitbit | now do you feel bad? | 20:22 |
| imsplitbit | none of that really happened | 20:22 |
| imsplitbit | I just wanted to make you feel bad | 20:22 |
| imsplitbit | moving on | 20:22 |
| imsplitbit | so I don't dislike the idea of every instance being an arbitrary thing that has at least one sub resource | 20:23 |
| imsplitbit | but we still get into issues of nested relationships | 20:23 |
| imsplitbit | WRT replication | 20:23 |
| vipul | Yea, it also becomes much like the cluster API we had | 20:27 |
| vipul | but what I don't really like at this point is why/how it's different | 20:27 |
| vipul | I mean what is it about clusters that makes it a differnt API than master/slave | 20:28 |
| *** amcrn has joined #openstack-trove | 20:28 | |
| vipul | I don't think it's a single endpoint type of thing.. I think that would be a good distinction if it was | 20:28 |
| vipul | but it seems like we have cases where we ant to know the IP of every 'node' | 20:28 |
| imsplitbit | agreed | 20:29 |
| yogesh | cluster is a single unit i guess, to be packaged as an instance......but replication is not... | 20:29 |
| vipul | the other distinction could be homogenious flavors | 20:29 |
| imsplitbit | yogesh: correct | 20:29 |
| yogesh | but there is another point of view... | 20:29 |
| yogesh | in essence it still is a single db instance getting provisioned.. | 20:29 |
| imsplitbit | well... not necessarily | 20:30 |
| imsplitbit | but the base case it is | 20:30 |
| yogesh | what cases, it is not? | 20:30 |
| yogesh | if there are such cases then probably god food for thought... | 20:30 |
| imsplitbit | I have a 2GB master that is replicating it's data to a 8GB slave that I also use for 50 other databases | 20:31 |
| imsplitbit | I see clustering different | 20:31 |
| imsplitbit | replication is *ONLY* a way of making sure data in one location also exists in another | 20:31 |
| yogesh | agree.. | 20:31 |
| imsplitbit | it shouldn't prescribe use case thoug | 20:31 |
| imsplitbit | though | 20:31 |
| hub_cap | ya u might only want to replicate _a_ table eh? | 20:32 |
| hub_cap | or replicating a single db in teh instance... | 20:33 |
| imsplitbit | the power we'll give our users is in saying "given that the service type you are provisioning has the replication capability enabled, you can now do some super cool stuff and we aren't gonna tell you how to do it" | 20:33 |
| imsplitbit | right hub_cap | 20:33 |
| yogesh | this is actually cross boundary replication... | 20:33 |
| yogesh | within the self contained group of nodes...if there is intra group replication that is mostly like clustering but inter-group configurations are not | 20:34 |
| imsplitbit | don't get me wrong, I don't dislike the sub resource node thing. but I don't think it lends itself to being open for doing interesting things with replication | 20:34 |
| imsplitbit | and I may be only representing the mysql pov here so please logic check me but I've seen some really interesting things done with replication that don't fit into a simple master-slave or master-master model | 20:35 |
| imsplitbit | I mean at their base, all relationships in mysql replication are master-slave, I would argue that all replication boils down to that | 20:36 |
| imsplitbit | but I want to make sure we represent that in the most concise way to our users | 20:36 |
| *** shivamshukla_ has quit IRC | 20:37 | |
| imsplitbit | and what governs this top level instance thing that may be one or more nodes? | 20:37 |
| imsplitbit | doesn't this break our current instance models? | 20:37 |
| yogesh | how abot not really decorating it as a cluster or replication | 20:37 |
| yogesh | leaving it at the group-of-nodes level | 20:38 |
| yogesh | from trove perspective | 20:38 |
| yogesh | and C or R as some kind of extensions... | 20:39 |
| *** dmakogon_ipod has quit IRC | 20:40 | |
| *** giulivo has joined #openstack-trove | 20:40 | |
| *** giulivo has left #openstack-trove | 20:40 | |
| kevinconway | imsplitbit: what if i want to do slave-slave-slave replication | 20:40 |
| *** ioggstream has joined #openstack-trove | 20:40 | |
| kevinconway | where nothing ever gets written but everything is replicated | 20:40 |
| ioggstream | hi@all | 20:40 |
| * imsplitbit /ignores kevinconway | 20:40 | |
| imsplitbit | :) | 20:40 |
| kevinconway | imsplitbit: i agree that primary-secondary or primary-primary hits 99% of all scenarios | 20:41 |
| imsplitbit | well even primary-primary is just primary-secondary bidirectionally | 20:41 |
| *** dmakogon_ipod has joined #openstack-trove | 20:42 | |
| imsplitbit | each one is a secondary to the other | 20:42 |
| kevinconway | true | 20:42 |
| imsplitbit | I just take issue with representing a group of N instances that are replicating to or from one another as one single thing with node subresources because representing the relationships difficult, which are paramount in replication | 20:43 |
| imsplitbit | and I'm sure thats because I haven't really seen how we would propose to represent the relationships this way | 20:43 |
| imsplitbit | without doing a ton of nasty nesting | 20:44 |
| dmakogon_ipod | vipul: ping ? | 20:44 |
| vipul | sorry here but in another meeting | 20:44 |
| kevinconway | yeah, i feel like we talked a lot about how to model the replica sets without deciding anything substantial | 20:45 |
| imsplitbit | ok well vipul and all, please ponder this. I'm writing up a new spec here https://wiki.openstack.org/wiki/Trove-Replication-And-Clustering-API | 20:45 |
| imsplitbit | kevinconway: I agree | 20:46 |
| imsplitbit | I think it makes sense 100% to represent clusters this way | 20:46 |
| kevinconway | i was fond of the idea of having instances as instances and adding a new resource to manage their relationships | 20:46 |
| imsplitbit | because they have a singular function. unlike replication which may or may not be a tool of just shuffling off copies of data here or there | 20:46 |
| imsplitbit | well I agree though that it makes the api look cleaner to keep everything an instance | 20:47 |
| imsplitbit | and ideally represent replication as a capability of a service type without hanging a whole other api path off of our default api | 20:47 |
| imsplitbit | then the only thing necessary to pull off replication is to extend the instance object with metadata to represent what it is and what it's relationship is | 20:48 |
| dmakogon_ipod | imsplitbit: could you make some short conclusion about storing models for clustering api | 20:49 |
| imsplitbit | where things fall apart is when you say I've got this replication thing that could grow into 4 tiers or more of parent-child relationships | 20:49 |
| imsplitbit | dmakogon_ipod: my proposal was to leave an instance an instance and add a metadata attribute to the returned ref that contains 2 things | 20:50 |
| imsplitbit | 1. replication_type: master-slave, master-master | 20:50 |
| dmakogon_ipod | imsplitbit: seems logicaly, almost the same as it was planned | 20:50 |
| imsplitbit | 2. replication_data: (maybe a bad name, config) relationship, i.e. where is it's master | 20:51 |
| imsplitbit | dmakogon_ipod: yes but I think the previous plan was to represent clusters and replication this way | 20:51 |
| imsplitbit | and I think making clusters into an instance with subresources is a cleaner way of displaying that | 20:51 |
| dmakogon_ipod | btw, how master-master replication organized ? | 20:51 |
| imsplitbit | well the logic there is really that it's just a master-slave relationship with multiple nodes replicating to one another | 20:52 |
| dmakogon_ipod | ok, than please add new replication type | 20:52 |
| dmakogon_ipod | for cassandra | 20:53 |
| imsplitbit | so you could probably simplify it by returning metadata that just says "replication_config: "master-ref"" | 20:53 |
| imsplitbit | sure | 20:53 |
| dmakogon_ipod | thanks | 20:54 |
| dmakogon_ipod | last question | 20:54 |
| imsplitbit | I've got about 20 minutes and I need to go pick up my daughter from school but I can continue this discussion later and/or tomorrow morning | 20:54 |
| dmakogon_ipod | does metadata would be extendable ? | 20:54 |
| imsplitbit | that is the intention | 20:55 |
| imsplitbit | we discussed this some time ago. | 20:55 |
| imsplitbit | and it seems to me that metadata should be useful for more than just replication | 20:55 |
| imsplitbit | and I would hate to extend the instance object with a different attribute for every possible capability of each service type | 20:56 |
| imsplitbit | the metatdata structure seems ideal for that | 20:56 |
| *** Barker has quit IRC | 20:56 | |
| imsplitbit | what are you thoughts on that dmakogon_ipod | 20:56 |
| imsplitbit | ? | 20:56 |
| yogesh | sorry, i missed some conversation but meanwhile was thinking about a topology with both clustering and replication...thoughts... | 20:56 |
| dmakogon_ipod | imsplitbit: i'm thinking about adding some service type related info | 20:57 |
| imsplitbit | dmakogon_ipod: like? | 20:57 |
| dmakogon_ipod | do not know what it is yet, but i'll try to bring something | 20:57 |
| imsplitbit | yogesh: I think cassandra has some very real use cases for describing topology in clustering | 20:57 |
| imsplitbit | mysql doesn't really need it | 20:57 |
| imsplitbit | cool dmakogon_ipod | 20:58 |
| imsplitbit | I think we need to really look at that | 20:58 |
| dmakogon_ipod | topology in very important | 20:58 |
| imsplitbit | because what we're proposing here is to change clusters to be represented as an instance with sub resources called nodes which are containers/vms in the background | 20:58 |
| imsplitbit | dmakogon_ipod: as I understand it with cassandra toplogy is very important | 20:59 |
| dmakogon_ipod | but in cassandra we could change it really easy, but what about mysql ? | 20:59 |
| dmakogon_ipod | imsplitbit: yes | 20:59 |
| dmakogon_ipod | imsplitbit: topology describes data flows between nodes in cluster | 20:59 |
| imsplitbit | well we'll just say that topology wrt clustering in mysql isn't too important to disply but is important for configuration. users may want to know but don't necessarily need to know | 21:00 |
| imsplitbit | but unfortunately for all of us mysql has several ways of being clustered | 21:00 |
| dmakogon_ipod | imsplitbit: so it could be optional)) | 21:00 |
| imsplitbit | possibly | 21:00 |
| imsplitbit | having been a DBA for several years I find topology VERY important no matter what I'm doing | 21:01 |
| *** Nate1 has quit IRC | 21:01 | |
| *** tanisdl has quit IRC | 21:01 | |
| imsplitbit | regarding what my data is being stored on | 21:01 |
| imsplitbit | dmakogon_ipod: please contribute usecases for cassandra to our open discussion on the etherpad. especially where topology requirements are different from mysql or redis | 21:02 |
| dmakogon_ipod | imsplitbit: you should go, your daughter waitinig for you)) | 21:02 |
| imsplitbit | https://etherpad.openstack.org/TroveClusteringAPI-WIP | 21:02 |
| dmakogon_ipod | ok, i'll do | 21:02 |
| imsplitbit | I've got about 10 minutes before I have to leave | 21:02 |
| dmakogon_ipod | imsplitbit: could i do it tomorrow ? | 21:03 |
| ioggstream | how long will it take to define these clustering api? | 21:03 |
| hub_cap | ioggstream: we are trying to define it now :) | 21:03 |
| imsplitbit | dmakogon_ipod: sure | 21:03 |
| hub_cap | we are pretty close though | 21:03 |
| imsplitbit | I'd like to discuss this more tomorrow and come to a conclusion pretty soon | 21:03 |
| dmakogon_ipod | imsplitbit: thanks | 21:03 |
| *** erkules_ is now known as erkules | 21:04 | |
| imsplitbit | because it either means that I need to completely stop work on the current impl of the clustering api and switch gears or get back to it and scrap all these new discussions | 21:04 |
| dmakogon_ipod | hub_cap: i have something to tell about current heat template | 21:04 |
| ioggstream | I just stumbled upon trove and guessing how much work will it take to make it production-ready | 21:04 |
| imsplitbit | ioggstream: depending on your definition it is | 21:04 |
| imsplitbit | we are running it in production already :) | 21:04 |
| dmakogon_ipod | imsplitbit: with something above it))) | 21:05 |
| ioggstream | imsplitbit: good to know! | 21:05 |
| dmakogon_ipod | hub_cap: ? | 21:05 |
| ioggstream | can you point me with some reference? | 21:05 |
| dmakogon_ipod | ioggstream: what do you want to see ? | 21:06 |
| ioggstream | giving reasons to my boss to let me develop on trove ;) | 21:06 |
| hub_cap | ioggstream: welcome! | 21:07 |
| imsplitbit | ioggstream: it's running here for sure http://www.rackspace.com/cloud/databases/ | 21:07 |
| imsplitbit | :) | 21:07 |
| ioggstream | I have some experience with python, mysql and ndb | 21:07 |
| yogesh | dmakogon: imsplitbit: so, if i define a cluster and i also want a replicated node (not as a part of the cluster) but separate...does it exist the part of the INSTANCE...if yes...how does it get handled in our design... | 21:07 |
| dmakogon_ipod | ioggstream: reason is very easy, developing Trove (OS DBaaS) is rather cheap than Amazon RDBS | 21:07 |
| *** adrian_otto has quit IRC | 21:07 | |
| ioggstream | dmakogon_ipod: I played with amz and its' interface is quite simple | 21:07 |
| dmakogon_ipod | ioggstream: but you can't still go deeper in it | 21:08 |
| dmakogon_ipod | ioggstream: only API | 21:08 |
| *** Nate1 has joined #openstack-trove | 21:08 | |
| ioggstream | dmakogon_ipod: yes, for some people it's a plus | 21:08 |
| dmakogon_ipod | yogesh: not clear to me, example ? | 21:08 |
| dmakogon_ipod | ioggstream: housewifes ? | 21:09 |
| imsplitbit | yogesh: I am not sure in the instance model if that works. replication is a capability of a service type so if you deploy galera cluster with 5 nodes you would have to replicate to another instance of service type galera cluster no? hub_cap chime in here | 21:09 |
| imsplitbit | I'm very sorry but I have to leave for a bit | 21:09 |
| dmakogon_ipod | k | 21:09 |
| imsplitbit | I'll log back in once I get home to continute this discussion | 21:09 |
| dmakogon_ipod | good converastion | 21:09 |
| ioggstream | dmakogon_ipod: don't underestimate housewifes | 21:09 |
| imsplitbit | very | 21:09 |
| ioggstream | imsplitbit: thx for your hints | 21:10 |
| *** pdmars has quit IRC | 21:10 | |
| ioggstream | so boys, it's 23.10 here in Italy, see you tomorrow: I'll try the https://wiki.openstack.org/wiki/Trove/installation | 21:11 |
| yogesh | dmakogon: if i want to create a cluster and want one of the nodes in the cluster to be the master of the replication group with one single slave for an instance...does that sound like a use case... | 21:11 |
| yogesh | and should we have trove define this combination of cluster/replication group | 21:11 |
| dmakogon_ipod | yogesh: seems like not so easy task to change cluster into replication group | 21:12 |
| ioggstream | nite | 21:13 |
| yogesh | lemme write something in detail... | 21:13 |
| yogesh | will post.. | 21:13 |
| yogesh | a gist... | 21:13 |
| dmakogon_ipod | ok, seems like i'm missing something | 21:13 |
| dmakogon_ipod | sorry | 21:13 |
| yogesh | dmakogon: not really...i myself am thinking aloud... | 21:14 |
| yogesh | pretty abstract,,, | 21:14 |
| yogesh | let me get my act together... ;-) | 21:14 |
| dmakogon_ipod | щл | 21:15 |
| dmakogon_ipod | ok | 21:15 |
| dmakogon_ipod | yogesh: btw, what's the status of templates ? | 21:16 |
| *** amcrn has quit IRC | 21:17 | |
| *** ioggstream has quit IRC | 21:17 | |
| yogesh | started...u should see something to review pretty soon | 21:17 |
| *** adrian_otto has joined #openstack-trove | 21:17 | |
| dmakogon_ipod | yogesh:with love))) | 21:19 |
| dmakogon_ipod | yogesh: i've got some plans for extending basic mysql template dynamically | 21:19 |
| dmakogon_ipod | volumes, security groups | 21:19 |
| dmakogon_ipod | floating ips | 21:19 |
| yogesh | dmakogon: dynamically? | 21:19 |
| *** mmcdaris has quit IRC | 21:20 | |
| *** fandikurnia01 has quit IRC | 21:20 | |
| dmakogon_ipod | yogesh: yes | 21:21 |
| dmakogon_ipod | if support is unabled: add specific part | 21:21 |
| dmakogon_ipod | should be easy | 21:21 |
| yogesh | ok...wud luv to see a tiny example | 21:21 |
| dmakogon_ipod | no problem | 21:22 |
| dmakogon_ipod | i could post it right now | 21:22 |
| yogesh | thanks much | 21:22 |
| dmakogon_ipod | https://github.com/openstack/trove/blob/master/trove/common/template.py - basic template | 21:22 |
| dmakogon_ipod | parsable by yaml lib | 21:23 |
| yogesh | ok | 21:23 |
| *** jasonb365 has joined #openstack-trove | 21:23 | |
| dmakogon_ipod | security group as extension: https://github.com/openstack/heat-templates/blob/master/cfn/deb/MultiNode_DevStack.yaml#L96-L104 | 21:24 |
| yogesh | so u mean abstracting the security groups out and pushing them into the template at run time... | 21:25 |
| yogesh | yeah thats fine.. | 21:26 |
| dmakogon_ipod | нуы | 21:26 |
| dmakogon_ipod | yes! | 21:26 |
| yogesh | but template in itself is dynamic strategy.. | 21:26 |
| yogesh | why to touch it at runtime.. | 21:26 |
| yogesh | let it be created at design time, full and consumable.. | 21:27 |
| *** isviridov has joined #openstack-trove | 21:27 | |
| *** mmcdaris has joined #openstack-trove | 21:27 | |
| dmakogon_ipod | floating ip: https://github.com/openstack/heat-templates/blob/a2ad92fa5897fb54645c7f572cd3467b3c31ff9e/cfn/F17/WordPress_2_Instances_With_EBS_EIP.template#L103-L109 | 21:27 |
| dmakogon_ipod | if to stay tottaly honest: https://github.com/openstack/heat-templates/blob/a2ad92fa5897fb54645c7f572cd3467b3c31ff9e/cfn/F17/WordPress_2_Instances_With_EBS_EIP.template#L98-L109 | 21:28 |
| dmakogon_ipod | yogesh: but i suppose you are ok with it | 21:28 |
| yogesh | well.. | 21:30 |
| yogesh | yeah...it depends | 21:31 |
| dmakogon_ipod | yogesh: depends on what ? | 21:32 |
| yogesh | dmakogon: on what exactly do we want to make dynamic...may be case to case... | 21:33 |
| dmakogon_ipod | yogesh: does heat is integrated to devstack ? | 21:33 |
| yogesh | yes | 21:33 |
| *** isviridov has quit IRC | 21:33 | |
| *** ashestakov_ has quit IRC | 21:33 | |
| yogesh | https://wiki.openstack.org/wiki/Heat/GettingStartedUsingDevstack | 21:33 |
| yogesh | :-) i did not get the question... | 21:34 |
| dmakogon_ipod | yogesh: i'm gonna do this because there is a unfilled gap between heat provisioning and nova provisioning in trove | 21:34 |
| *** Nate1 has quit IRC | 21:34 | |
| yogesh | oh ok... | 21:34 |
| *** openstackgerrit has quit IRC | 21:34 | |
| yogesh | i got u.. | 21:34 |
| dmakogon_ipod | let me show an example | 21:35 |
| yogesh | so u want to handle the case when heat is not enabled... | 21:35 |
| dmakogon_ipod | yes | 21:35 |
| yogesh | hmm | 21:35 |
| yogesh | yeah, sure...an example.. | 21:35 |
| yogesh | thanks | 21:35 |
| *** tanisdl has joined #openstack-trove | 21:36 | |
| *** djohnstone has joined #openstack-trove | 21:36 | |
| dmakogon_ipod | user creating instance, request comes into trove-api (security group is created and assigned to trove-instance) | 21:36 |
| dmakogon_ipod | than, request comes to taskmanager, and going to heat provisioning func | 21:37 |
| dmakogon_ipod | stack, resources are created | 21:37 |
| *** amcrn has joined #openstack-trove | 21:37 | |
| dmakogon_ipod | but (!!!) security not assigned to compute instance | 21:37 |
| dmakogon_ipod | even security group parameter doesn't used | 21:38 |
| dmakogon_ipod | so, SG is a gap #1 | 21:38 |
| *** jlucci has joined #openstack-trove | 21:39 | |
| yogesh | agreed... | 21:39 |
| dmakogon_ipod | so, here comes a bug | 21:39 |
| dmakogon_ipod | trove-cli security group delete with heat support would cause an exception | 21:40 |
| *** jlucci has quit IRC | 21:40 | |
| yogesh | makes sense... | 21:40 |
| dmakogon_ipod | oh, no | 21:40 |
| dmakogon_ipod | it wouldn't | 21:40 |
| yogesh | it would be a bug but wont cause exception | 21:41 |
| *** isviridov has joined #openstack-trove | 21:41 | |
| dmakogon_ipod | because security groups are deleting directly with novaclient.security_groups.delete(...) | 21:42 |
| *** demorris has quit IRC | 21:42 | |
| yogesh | i wanted to figure if trove has security-group management apis.. | 21:42 |
| yogesh | it hasn't | 21:43 |
| dmakogon_ipod | management? - np | 21:43 |
| dmakogon_ipod | only user API | 21:43 |
| dmakogon_ipod | in some way it is strange to me | 21:43 |
| dmakogon_ipod | if i am trove user and i want only a database why should a be able to manage that kind of resource ? | 21:44 |
| yogesh | yeah | 21:45 |
| yogesh | dmkogon: need to leave for a bit | 21:46 |
| yogesh | will be back in few minutes | 21:46 |
| dmakogon_ipod | i'm going to sleep | 21:47 |
| yogesh | didn't rezlize its late for u | 21:47 |
| yogesh | realize* | 21:48 |
| *** yogeshmehra has joined #openstack-trove | 21:48 | |
| *** yogeshmehra has left #openstack-trove | 21:49 | |
| *** haomaiwa_ has joined #openstack-trove | 21:49 | |
| *** haomaiwang has quit IRC | 21:49 | |
| *** yogesh has quit IRC | 21:52 | |
| hub_cap | haha yup | 21:55 |
| *** grapex has quit IRC | 21:58 | |
| *** dmakogon_ipod has quit IRC | 22:00 | |
| *** yogesh has joined #openstack-trove | 22:00 | |
| *** yogeshmehra has joined #openstack-trove | 22:05 | |
| *** yogeshmehra has quit IRC | 22:10 | |
| *** amytron has quit IRC | 22:15 | |
| *** adrian_otto has quit IRC | 22:16 | |
| *** robertmyers has quit IRC | 22:17 | |
| *** jcru has quit IRC | 22:24 | |
| *** jlucci has joined #openstack-trove | 22:24 | |
| *** jlucci has quit IRC | 22:25 | |
| *** yogesh has quit IRC | 22:25 | |
| *** amytron has joined #openstack-trove | 22:28 | |
| *** jasonb365 has quit IRC | 22:29 | |
| *** jlucci has joined #openstack-trove | 22:29 | |
| *** amytron has quit IRC | 22:31 | |
| *** rnirmal has quit IRC | 22:32 | |
| *** amytron has joined #openstack-trove | 22:45 | |
| *** mmcdaris has quit IRC | 22:50 | |
| *** jlucci has quit IRC | 22:59 | |
| *** openstackgerrit has joined #openstack-trove | 23:01 | |
| *** jlucci has joined #openstack-trove | 23:03 | |
| *** dafter has quit IRC | 23:05 | |
| *** tanisdl has quit IRC | 23:09 | |
| *** tanisdl has joined #openstack-trove | 23:09 | |
| *** paul_lodronio has quit IRC | 23:11 | |
| openstackgerrit | Justin Hopper proposed a change to openstack/trove: Fix Timestamps for Resize Usage Events https://review.openstack.org/50192 | 23:23 |
| *** mmcdaris has joined #openstack-trove | 23:24 | |
| *** jlucci has quit IRC | 23:27 | |
| *** amytron has quit IRC | 23:29 | |
| *** adrian_otto has joined #openstack-trove | 23:31 | |
| juice | hub_cap: this patch (^ ^) will fix the intermittent false negative build failures | 23:37 |
| *** adrian_otto has quit IRC | 23:37 | |
| hub_cap | ++ | 23:38 |
| hub_cap | haha juice youve done this before ;) | 23:38 |
| juice | oh yes I have | 23:39 |
| juice | but I missed the resize event the last time | 23:39 |
| hub_cap | lulz | 23:39 |
| juice | I think we have it all covered now | 23:39 |
| hub_cap | once the robots +1 have vipul or SlickNik review it | 23:39 |
| juice | good riddance I say - good riddance | 23:39 |
| hub_cap | SlickNik: lol at you +1'ing the 'reap' thing in nova super fast | 23:39 |
| hub_cap | juice: are u shaking your fist? | 23:40 |
| hub_cap | https://wiki.openstack.org/wiki/Testr#Debugging_.28pdb.29_Tests | 23:40 |
| hub_cap | ^ ^ so nice | 23:40 |
| SlickNik | hahah, can you tell I've been bitten by it :P | 23:40 |
| hub_cap | wouldve NEVER guessed | 23:40 |
| hub_cap | :P | 23:40 |
| juice | with esp and slicknik as my witnesses | 23:40 |
| hub_cap | im so glad i can pdb my shit now | 23:40 |
| hub_cap | lol juice | 23:40 |
| *** tanisdl has quit IRC | 23:41 | |
| hub_cap | sweet found my bug thanks to pdb | 23:45 |
| *** djohnstone has quit IRC | 23:53 | |
| hub_cap | ok im out for the day | 23:56 |
| hub_cap | l8r all | 23:57 |
| juice | c ya hub_cap | 23:57 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
